scannerc.go 80 KB

12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758596061626364656667686970717273747576777879808182838485868788899091929394959697989910010110210310410510610710810911011111211311411511611711811912012112212312412512612712812913013113213313413513613713813914014114214314414514614714814915015115215315415515615715815916016116216316416516616716816917017117217317417517617717817918018118218318418518618718818919019119219319419519619719819920020120220320420520620720820921021121221321421521621721821922022122222322422522622722822923023123223323423523623723823924024124224324424524624724824925025125225325425525625725825926026126226326426526626726826927027127227327427527627727827928028128228328428528628728828929029129229329429529629729829930030130230330430530630730830931031131231331431531631731831932032132232332432532632732832933033133233333433533633733833934034134234334434534634734834935035135235335435535635735835936036136236336436536636736836937037137237337437537637737837938038138238338438538638738838939039139239339439539639739839940040140240340440540640740840941041141241341441541641741841942042142242342442542642742842943043143243343443543643743843944044144244344444544644744844945045145245345445545645745845946046146246346446546646746846947047147247347447547647747847948048148248348448548648748848949049149249349449549649749849950050150250350450550650750850951051151251351451551651751851952052152252352452552652752852953053153253353453553653753853954054154254354454554654754854955055155255355455555655755855956056156256356456556656756856957057157257357457557657757857958058158258358458558658758858959059159259359459559659759859960060160260360460560660760860961061161261361461561661761861962062162262362462562662762862963063163263363463563663763863964064164264364464564664764864965065165265365465565665765865966066166266366466566666766866967067167267367467567667767867968068168268368468568668768868969069169269369469569669769869970070170270370470570670770870971071171271371471571671771871972072172272372472572672772872973073173273373473573673773873974074174274374474574674774874975075175275375475575675775875976076176276376476576676776876977077177277377477577677777877978078178278378478578678778878979079179279379479579679779879980080180280380480580680780880981081181281381481581681781881982082182282382482582682782882983083183283383483583683783883984084184284384484584684784884985085185285385485585685785885986086186286386486586686786886987087187287387487587687787887988088188288388488588688788888989089189289389489589689789889990090190290390490590690790890991091191291391491591691791891992092192292392492592692792892993093193293393493593693793893994094194294394494594694794894995095195295395495595695795895996096196296396496596696796896997097197297397497597697797897998098198298398498598698798898999099199299399499599699799899910001001100210031004100510061007100810091010101110121013101410151016101710181019102010211022102310241025102610271028102910301031103210331034103510361037103810391040104110421043104410451046104710481049105010511052105310541055105610571058105910601061106210631064106510661067106810691070107110721073107410751076107710781079108010811082108310841085108610871088108910901091109210931094109510961097109810991100110111021103110411051106110711081109111011111112111311141115111611171118111911201121112211231124112511261127112811291130113111321133113411351136113711381139114011411142114311441145114611471148114911501151115211531154115511561157115811591160116111621163116411651166116711681169117011711172117311741175117611771178117911801181118211831184118511861187118811891190119111921193119411951196119711981199120012011202120312041205120612071208120912101211121212131214121512161217121812191220122112221223122412251226122712281229123012311232123312341235123612371238123912401241124212431244124512461247124812491250125112521253125412551256125712581259126012611262126312641265126612671268126912701271127212731274127512761277127812791280128112821283128412851286128712881289129012911292129312941295129612971298129913001301130213031304130513061307130813091310131113121313131413151316131713181319132013211322132313241325132613271328132913301331133213331334133513361337133813391340134113421343134413451346134713481349135013511352135313541355135613571358135913601361136213631364136513661367136813691370137113721373137413751376137713781379138013811382138313841385138613871388138913901391139213931394139513961397139813991400140114021403140414051406140714081409141014111412141314141415141614171418141914201421142214231424142514261427142814291430143114321433143414351436143714381439144014411442144314441445144614471448144914501451145214531454145514561457145814591460146114621463146414651466146714681469147014711472147314741475147614771478147914801481148214831484148514861487148814891490149114921493149414951496149714981499150015011502150315041505150615071508150915101511151215131514151515161517151815191520152115221523152415251526152715281529153015311532153315341535153615371538153915401541154215431544154515461547154815491550155115521553155415551556155715581559156015611562156315641565156615671568156915701571157215731574157515761577157815791580158115821583158415851586158715881589159015911592159315941595159615971598159916001601160216031604160516061607160816091610161116121613161416151616161716181619162016211622162316241625162616271628162916301631163216331634163516361637163816391640164116421643164416451646164716481649165016511652165316541655165616571658165916601661166216631664166516661667166816691670167116721673167416751676167716781679168016811682168316841685168616871688168916901691169216931694169516961697169816991700170117021703170417051706170717081709171017111712171317141715171617171718171917201721172217231724172517261727172817291730173117321733173417351736173717381739174017411742174317441745174617471748174917501751175217531754175517561757175817591760176117621763176417651766176717681769177017711772177317741775177617771778177917801781178217831784178517861787178817891790179117921793179417951796179717981799180018011802180318041805180618071808180918101811181218131814181518161817181818191820182118221823182418251826182718281829183018311832183318341835183618371838183918401841184218431844184518461847184818491850185118521853185418551856185718581859186018611862186318641865186618671868186918701871187218731874187518761877187818791880188118821883188418851886188718881889189018911892189318941895189618971898189919001901190219031904190519061907190819091910191119121913191419151916191719181919192019211922192319241925192619271928192919301931193219331934193519361937193819391940194119421943194419451946194719481949195019511952195319541955195619571958195919601961196219631964196519661967196819691970197119721973197419751976197719781979198019811982198319841985198619871988198919901991199219931994199519961997199819992000200120022003200420052006200720082009201020112012201320142015201620172018201920202021202220232024202520262027202820292030203120322033203420352036203720382039204020412042204320442045204620472048204920502051205220532054205520562057205820592060206120622063206420652066206720682069207020712072207320742075207620772078207920802081208220832084208520862087208820892090209120922093209420952096209720982099210021012102210321042105210621072108210921102111211221132114211521162117211821192120212121222123212421252126212721282129213021312132213321342135213621372138213921402141214221432144214521462147214821492150215121522153215421552156215721582159216021612162216321642165216621672168216921702171217221732174217521762177217821792180218121822183218421852186218721882189219021912192219321942195219621972198219922002201220222032204220522062207220822092210221122122213221422152216221722182219222022212222222322242225222622272228222922302231223222332234223522362237223822392240224122422243224422452246224722482249225022512252225322542255225622572258225922602261226222632264226522662267226822692270227122722273227422752276227722782279228022812282228322842285228622872288228922902291229222932294229522962297229822992300230123022303230423052306230723082309231023112312231323142315231623172318231923202321232223232324232523262327232823292330233123322333233423352336233723382339234023412342234323442345234623472348234923502351235223532354235523562357235823592360236123622363236423652366236723682369237023712372237323742375237623772378237923802381238223832384238523862387238823892390239123922393239423952396239723982399240024012402240324042405240624072408240924102411241224132414241524162417241824192420242124222423242424252426242724282429243024312432243324342435243624372438243924402441244224432444244524462447244824492450245124522453245424552456245724582459246024612462246324642465246624672468246924702471247224732474247524762477247824792480248124822483248424852486248724882489249024912492249324942495249624972498249925002501250225032504250525062507250825092510251125122513251425152516251725182519252025212522252325242525252625272528252925302531253225332534253525362537253825392540254125422543254425452546254725482549255025512552255325542555255625572558255925602561256225632564256525662567256825692570257125722573257425752576257725782579258025812582258325842585258625872588258925902591259225932594259525962597259825992600260126022603260426052606260726082609261026112612261326142615261626172618261926202621262226232624262526262627262826292630263126322633263426352636263726382639264026412642264326442645264626472648264926502651265226532654265526562657265826592660266126622663266426652666266726682669267026712672267326742675267626772678267926802681268226832684268526862687268826892690269126922693269426952696269726982699270027012702270327042705270627072708270927102711271227132714271527162717271827192720272127222723272427252726272727282729273027312732273327342735273627372738273927402741274227432744274527462747274827492750275127522753275427552756275727582759276027612762276327642765276627672768276927702771277227732774277527762777277827792780278127822783278427852786278727882789279027912792279327942795279627972798279928002801280228032804280528062807280828092810281128122813281428152816281728182819282028212822282328242825282628272828282928302831283228332834283528362837283828392840284128422843284428452846284728482849285028512852285328542855285628572858285928602861286228632864286528662867286828692870287128722873
  1. package yaml
  2. import (
  3. "bytes"
  4. "fmt"
  5. )
  6. // Introduction
  7. // ************
  8. //
  9. // The following notes assume that you are familiar with the YAML specification
  10. // (http://yaml.org/spec/1.2/spec.html). We mostly follow it, although in
  11. // some cases we are less restrictive that it requires.
  12. //
  13. // The process of transforming a YAML stream into a sequence of events is
  14. // divided on two steps: Scanning and Parsing.
  15. //
  16. // The Scanner transforms the input stream into a sequence of tokens, while the
  17. // parser transform the sequence of tokens produced by the Scanner into a
  18. // sequence of parsing events.
  19. //
  20. // The Scanner is rather clever and complicated. The Parser, on the contrary,
  21. // is a straightforward implementation of a recursive-descendant parser (or,
  22. // LL(1) parser, as it is usually called).
  23. //
  24. // Actually there are two issues of Scanning that might be called "clever", the
  25. // rest is quite straightforward. The issues are "block collection start" and
  26. // "simple keys". Both issues are explained below in details.
  27. //
  28. // Here the Scanning step is explained and implemented. We start with the list
  29. // of all the tokens produced by the Scanner together with short descriptions.
  30. //
  31. // Now, tokens:
  32. //
  33. // STREAM-START(encoding) # The stream start.
  34. // STREAM-END # The stream end.
  35. // VERSION-DIRECTIVE(major,minor) # The '%YAML' directive.
  36. // TAG-DIRECTIVE(handle,prefix) # The '%TAG' directive.
  37. // DOCUMENT-START # '---'
  38. // DOCUMENT-END # '...'
  39. // BLOCK-SEQUENCE-START # Indentation increase denoting a block
  40. // BLOCK-MAPPING-START # sequence or a block mapping.
  41. // BLOCK-END # Indentation decrease.
  42. // FLOW-SEQUENCE-START # '['
  43. // FLOW-SEQUENCE-END # ']'
  44. // BLOCK-SEQUENCE-START # '{'
  45. // BLOCK-SEQUENCE-END # '}'
  46. // BLOCK-ENTRY # '-'
  47. // FLOW-ENTRY # ','
  48. // KEY # '?' or nothing (simple keys).
  49. // VALUE # ':'
  50. // ALIAS(anchor) # '*anchor'
  51. // ANCHOR(anchor) # '&anchor'
  52. // TAG(handle,suffix) # '!handle!suffix'
  53. // SCALAR(value,style) # A scalar.
  54. //
  55. // The following two tokens are "virtual" tokens denoting the beginning and the
  56. // end of the stream:
  57. //
  58. // STREAM-START(encoding)
  59. // STREAM-END
  60. //
  61. // We pass the information about the input stream encoding with the
  62. // STREAM-START token.
  63. //
  64. // The next two tokens are responsible for tags:
  65. //
  66. // VERSION-DIRECTIVE(major,minor)
  67. // TAG-DIRECTIVE(handle,prefix)
  68. //
  69. // Example:
  70. //
  71. // %YAML 1.1
  72. // %TAG ! !foo
  73. // %TAG !yaml! tag:yaml.org,2002:
  74. // ---
  75. //
  76. // The correspoding sequence of tokens:
  77. //
  78. // STREAM-START(utf-8)
  79. // VERSION-DIRECTIVE(1,1)
  80. // TAG-DIRECTIVE("!","!foo")
  81. // TAG-DIRECTIVE("!yaml","tag:yaml.org,2002:")
  82. // DOCUMENT-START
  83. // STREAM-END
  84. //
  85. // Note that the VERSION-DIRECTIVE and TAG-DIRECTIVE tokens occupy a whole
  86. // line.
  87. //
  88. // The document start and end indicators are represented by:
  89. //
  90. // DOCUMENT-START
  91. // DOCUMENT-END
  92. //
  93. // Note that if a YAML stream contains an implicit document (without '---'
  94. // and '...' indicators), no DOCUMENT-START and DOCUMENT-END tokens will be
  95. // produced.
  96. //
  97. // In the following examples, we present whole documents together with the
  98. // produced tokens.
  99. //
  100. // 1. An implicit document:
  101. //
  102. // 'a scalar'
  103. //
  104. // Tokens:
  105. //
  106. // STREAM-START(utf-8)
  107. // SCALAR("a scalar",single-quoted)
  108. // STREAM-END
  109. //
  110. // 2. An explicit document:
  111. //
  112. // ---
  113. // 'a scalar'
  114. // ...
  115. //
  116. // Tokens:
  117. //
  118. // STREAM-START(utf-8)
  119. // DOCUMENT-START
  120. // SCALAR("a scalar",single-quoted)
  121. // DOCUMENT-END
  122. // STREAM-END
  123. //
  124. // 3. Several documents in a stream:
  125. //
  126. // 'a scalar'
  127. // ---
  128. // 'another scalar'
  129. // ---
  130. // 'yet another scalar'
  131. //
  132. // Tokens:
  133. //
  134. // STREAM-START(utf-8)
  135. // SCALAR("a scalar",single-quoted)
  136. // DOCUMENT-START
  137. // SCALAR("another scalar",single-quoted)
  138. // DOCUMENT-START
  139. // SCALAR("yet another scalar",single-quoted)
  140. // STREAM-END
  141. //
  142. // We have already introduced the SCALAR token above. The following tokens are
  143. // used to describe aliases, anchors, tag, and scalars:
  144. //
  145. // ALIAS(anchor)
  146. // ANCHOR(anchor)
  147. // TAG(handle,suffix)
  148. // SCALAR(value,style)
  149. //
  150. // The following series of examples illustrate the usage of these tokens:
  151. //
  152. // 1. A recursive sequence:
  153. //
  154. // &A [ *A ]
  155. //
  156. // Tokens:
  157. //
  158. // STREAM-START(utf-8)
  159. // ANCHOR("A")
  160. // FLOW-SEQUENCE-START
  161. // ALIAS("A")
  162. // FLOW-SEQUENCE-END
  163. // STREAM-END
  164. //
  165. // 2. A tagged scalar:
  166. //
  167. // !!float "3.14" # A good approximation.
  168. //
  169. // Tokens:
  170. //
  171. // STREAM-START(utf-8)
  172. // TAG("!!","float")
  173. // SCALAR("3.14",double-quoted)
  174. // STREAM-END
  175. //
  176. // 3. Various scalar styles:
  177. //
  178. // --- # Implicit empty plain scalars do not produce tokens.
  179. // --- a plain scalar
  180. // --- 'a single-quoted scalar'
  181. // --- "a double-quoted scalar"
  182. // --- |-
  183. // a literal scalar
  184. // --- >-
  185. // a folded
  186. // scalar
  187. //
  188. // Tokens:
  189. //
  190. // STREAM-START(utf-8)
  191. // DOCUMENT-START
  192. // DOCUMENT-START
  193. // SCALAR("a plain scalar",plain)
  194. // DOCUMENT-START
  195. // SCALAR("a single-quoted scalar",single-quoted)
  196. // DOCUMENT-START
  197. // SCALAR("a double-quoted scalar",double-quoted)
  198. // DOCUMENT-START
  199. // SCALAR("a literal scalar",literal)
  200. // DOCUMENT-START
  201. // SCALAR("a folded scalar",folded)
  202. // STREAM-END
  203. //
  204. // Now it's time to review collection-related tokens. We will start with
  205. // flow collections:
  206. //
  207. // FLOW-SEQUENCE-START
  208. // FLOW-SEQUENCE-END
  209. // FLOW-MAPPING-START
  210. // FLOW-MAPPING-END
  211. // FLOW-ENTRY
  212. // KEY
  213. // VALUE
  214. //
  215. // The tokens FLOW-SEQUENCE-START, FLOW-SEQUENCE-END, FLOW-MAPPING-START, and
  216. // FLOW-MAPPING-END represent the indicators '[', ']', '{', and '}'
  217. // correspondingly. FLOW-ENTRY represent the ',' indicator. Finally the
  218. // indicators '?' and ':', which are used for denoting mapping keys and values,
  219. // are represented by the KEY and VALUE tokens.
  220. //
  221. // The following examples show flow collections:
  222. //
  223. // 1. A flow sequence:
  224. //
  225. // [item 1, item 2, item 3]
  226. //
  227. // Tokens:
  228. //
  229. // STREAM-START(utf-8)
  230. // FLOW-SEQUENCE-START
  231. // SCALAR("item 1",plain)
  232. // FLOW-ENTRY
  233. // SCALAR("item 2",plain)
  234. // FLOW-ENTRY
  235. // SCALAR("item 3",plain)
  236. // FLOW-SEQUENCE-END
  237. // STREAM-END
  238. //
  239. // 2. A flow mapping:
  240. //
  241. // {
  242. // a simple key: a value, # Note that the KEY token is produced.
  243. // ? a complex key: another value,
  244. // }
  245. //
  246. // Tokens:
  247. //
  248. // STREAM-START(utf-8)
  249. // FLOW-MAPPING-START
  250. // KEY
  251. // SCALAR("a simple key",plain)
  252. // VALUE
  253. // SCALAR("a value",plain)
  254. // FLOW-ENTRY
  255. // KEY
  256. // SCALAR("a complex key",plain)
  257. // VALUE
  258. // SCALAR("another value",plain)
  259. // FLOW-ENTRY
  260. // FLOW-MAPPING-END
  261. // STREAM-END
  262. //
  263. // A simple key is a key which is not denoted by the '?' indicator. Note that
  264. // the Scanner still produce the KEY token whenever it encounters a simple key.
  265. //
  266. // For scanning block collections, the following tokens are used (note that we
  267. // repeat KEY and VALUE here):
  268. //
  269. // BLOCK-SEQUENCE-START
  270. // BLOCK-MAPPING-START
  271. // BLOCK-END
  272. // BLOCK-ENTRY
  273. // KEY
  274. // VALUE
  275. //
  276. // The tokens BLOCK-SEQUENCE-START and BLOCK-MAPPING-START denote indentation
  277. // increase that precedes a block collection (cf. the INDENT token in Python).
  278. // The token BLOCK-END denote indentation decrease that ends a block collection
  279. // (cf. the DEDENT token in Python). However YAML has some syntax pecularities
  280. // that makes detections of these tokens more complex.
  281. //
  282. // The tokens BLOCK-ENTRY, KEY, and VALUE are used to represent the indicators
  283. // '-', '?', and ':' correspondingly.
  284. //
  285. // The following examples show how the tokens BLOCK-SEQUENCE-START,
  286. // BLOCK-MAPPING-START, and BLOCK-END are emitted by the Scanner:
  287. //
  288. // 1. Block sequences:
  289. //
  290. // - item 1
  291. // - item 2
  292. // -
  293. // - item 3.1
  294. // - item 3.2
  295. // -
  296. // key 1: value 1
  297. // key 2: value 2
  298. //
  299. // Tokens:
  300. //
  301. // STREAM-START(utf-8)
  302. // BLOCK-SEQUENCE-START
  303. // BLOCK-ENTRY
  304. // SCALAR("item 1",plain)
  305. // BLOCK-ENTRY
  306. // SCALAR("item 2",plain)
  307. // BLOCK-ENTRY
  308. // BLOCK-SEQUENCE-START
  309. // BLOCK-ENTRY
  310. // SCALAR("item 3.1",plain)
  311. // BLOCK-ENTRY
  312. // SCALAR("item 3.2",plain)
  313. // BLOCK-END
  314. // BLOCK-ENTRY
  315. // BLOCK-MAPPING-START
  316. // KEY
  317. // SCALAR("key 1",plain)
  318. // VALUE
  319. // SCALAR("value 1",plain)
  320. // KEY
  321. // SCALAR("key 2",plain)
  322. // VALUE
  323. // SCALAR("value 2",plain)
  324. // BLOCK-END
  325. // BLOCK-END
  326. // STREAM-END
  327. //
  328. // 2. Block mappings:
  329. //
  330. // a simple key: a value # The KEY token is produced here.
  331. // ? a complex key
  332. // : another value
  333. // a mapping:
  334. // key 1: value 1
  335. // key 2: value 2
  336. // a sequence:
  337. // - item 1
  338. // - item 2
  339. //
  340. // Tokens:
  341. //
  342. // STREAM-START(utf-8)
  343. // BLOCK-MAPPING-START
  344. // KEY
  345. // SCALAR("a simple key",plain)
  346. // VALUE
  347. // SCALAR("a value",plain)
  348. // KEY
  349. // SCALAR("a complex key",plain)
  350. // VALUE
  351. // SCALAR("another value",plain)
  352. // KEY
  353. // SCALAR("a mapping",plain)
  354. // BLOCK-MAPPING-START
  355. // KEY
  356. // SCALAR("key 1",plain)
  357. // VALUE
  358. // SCALAR("value 1",plain)
  359. // KEY
  360. // SCALAR("key 2",plain)
  361. // VALUE
  362. // SCALAR("value 2",plain)
  363. // BLOCK-END
  364. // KEY
  365. // SCALAR("a sequence",plain)
  366. // VALUE
  367. // BLOCK-SEQUENCE-START
  368. // BLOCK-ENTRY
  369. // SCALAR("item 1",plain)
  370. // BLOCK-ENTRY
  371. // SCALAR("item 2",plain)
  372. // BLOCK-END
  373. // BLOCK-END
  374. // STREAM-END
  375. //
  376. // YAML does not always require to start a new block collection from a new
  377. // line. If the current line contains only '-', '?', and ':' indicators, a new
  378. // block collection may start at the current line. The following examples
  379. // illustrate this case:
  380. //
  381. // 1. Collections in a sequence:
  382. //
  383. // - - item 1
  384. // - item 2
  385. // - key 1: value 1
  386. // key 2: value 2
  387. // - ? complex key
  388. // : complex value
  389. //
  390. // Tokens:
  391. //
  392. // STREAM-START(utf-8)
  393. // BLOCK-SEQUENCE-START
  394. // BLOCK-ENTRY
  395. // BLOCK-SEQUENCE-START
  396. // BLOCK-ENTRY
  397. // SCALAR("item 1",plain)
  398. // BLOCK-ENTRY
  399. // SCALAR("item 2",plain)
  400. // BLOCK-END
  401. // BLOCK-ENTRY
  402. // BLOCK-MAPPING-START
  403. // KEY
  404. // SCALAR("key 1",plain)
  405. // VALUE
  406. // SCALAR("value 1",plain)
  407. // KEY
  408. // SCALAR("key 2",plain)
  409. // VALUE
  410. // SCALAR("value 2",plain)
  411. // BLOCK-END
  412. // BLOCK-ENTRY
  413. // BLOCK-MAPPING-START
  414. // KEY
  415. // SCALAR("complex key")
  416. // VALUE
  417. // SCALAR("complex value")
  418. // BLOCK-END
  419. // BLOCK-END
  420. // STREAM-END
  421. //
  422. // 2. Collections in a mapping:
  423. //
  424. // ? a sequence
  425. // : - item 1
  426. // - item 2
  427. // ? a mapping
  428. // : key 1: value 1
  429. // key 2: value 2
  430. //
  431. // Tokens:
  432. //
  433. // STREAM-START(utf-8)
  434. // BLOCK-MAPPING-START
  435. // KEY
  436. // SCALAR("a sequence",plain)
  437. // VALUE
  438. // BLOCK-SEQUENCE-START
  439. // BLOCK-ENTRY
  440. // SCALAR("item 1",plain)
  441. // BLOCK-ENTRY
  442. // SCALAR("item 2",plain)
  443. // BLOCK-END
  444. // KEY
  445. // SCALAR("a mapping",plain)
  446. // VALUE
  447. // BLOCK-MAPPING-START
  448. // KEY
  449. // SCALAR("key 1",plain)
  450. // VALUE
  451. // SCALAR("value 1",plain)
  452. // KEY
  453. // SCALAR("key 2",plain)
  454. // VALUE
  455. // SCALAR("value 2",plain)
  456. // BLOCK-END
  457. // BLOCK-END
  458. // STREAM-END
  459. //
  460. // YAML also permits non-indented sequences if they are included into a block
  461. // mapping. In this case, the token BLOCK-SEQUENCE-START is not produced:
  462. //
  463. // key:
  464. // - item 1 # BLOCK-SEQUENCE-START is NOT produced here.
  465. // - item 2
  466. //
  467. // Tokens:
  468. //
  469. // STREAM-START(utf-8)
  470. // BLOCK-MAPPING-START
  471. // KEY
  472. // SCALAR("key",plain)
  473. // VALUE
  474. // BLOCK-ENTRY
  475. // SCALAR("item 1",plain)
  476. // BLOCK-ENTRY
  477. // SCALAR("item 2",plain)
  478. // BLOCK-END
  479. //
  480. // Ensure that the buffer contains the required number of characters.
  481. // Return true on success, false on failure (reader error or memory error).
  482. func cache(parser *yaml_parser_t, length int) bool {
  483. // [Go] This was inlined: !cache(A, B) -> unread < B && !update(A, B)
  484. return parser.unread >= length || yaml_parser_update_buffer(parser, length)
  485. }
  486. // Advance the buffer pointer.
  487. func skip(parser *yaml_parser_t) {
  488. parser.mark.index++
  489. parser.mark.column++
  490. parser.unread--
  491. parser.buffer_pos += width(parser.buffer[parser.buffer_pos])
  492. }
  493. func skip_line(parser *yaml_parser_t) {
  494. if is_crlf(parser.buffer, parser.buffer_pos) {
  495. parser.mark.index += 2
  496. parser.mark.column = 0
  497. parser.mark.line++
  498. parser.unread -= 2
  499. parser.buffer_pos += 2
  500. } else if is_break(parser.buffer, parser.buffer_pos) {
  501. parser.mark.index++
  502. parser.mark.column = 0
  503. parser.mark.line++
  504. parser.unread--
  505. parser.buffer_pos += width(parser.buffer[parser.buffer_pos])
  506. }
  507. }
  508. // Copy a character to a string buffer and advance pointers.
  509. func read(parser *yaml_parser_t, s []byte) []byte {
  510. w := width(parser.buffer[parser.buffer_pos])
  511. if w == 0 {
  512. panic("invalid character sequence")
  513. }
  514. if len(s) == 0 {
  515. s = make([]byte, 0, 32)
  516. }
  517. if w == 1 && len(s)+w <= cap(s) {
  518. s = s[:len(s)+1]
  519. s[len(s)-1] = parser.buffer[parser.buffer_pos]
  520. parser.buffer_pos++
  521. } else {
  522. s = append(s, parser.buffer[parser.buffer_pos:parser.buffer_pos+w]...)
  523. parser.buffer_pos += w
  524. }
  525. parser.mark.index++
  526. parser.mark.column++
  527. parser.unread--
  528. return s
  529. }
  530. // Copy a line break character to a string buffer and advance pointers.
  531. func read_line(parser *yaml_parser_t, s []byte) []byte {
  532. buf := parser.buffer
  533. pos := parser.buffer_pos
  534. switch {
  535. case buf[pos] == '\r' && buf[pos+1] == '\n':
  536. // CR LF . LF
  537. s = append(s, '\n')
  538. parser.buffer_pos += 2
  539. parser.mark.index++
  540. parser.unread--
  541. case buf[pos] == '\r' || buf[pos] == '\n':
  542. // CR|LF . LF
  543. s = append(s, '\n')
  544. parser.buffer_pos += 1
  545. case buf[pos] == '\xC2' && buf[pos+1] == '\x85':
  546. // NEL . LF
  547. s = append(s, '\n')
  548. parser.buffer_pos += 2
  549. case buf[pos] == '\xE2' && buf[pos+1] == '\x80' && (buf[pos+2] == '\xA8' || buf[pos+2] == '\xA9'):
  550. // LS|PS . LS|PS
  551. s = append(s, buf[parser.buffer_pos:pos+3]...)
  552. parser.buffer_pos += 3
  553. default:
  554. return s
  555. }
  556. parser.mark.index++
  557. parser.mark.column = 0
  558. parser.mark.line++
  559. parser.unread--
  560. return s
  561. }
  562. // Get the next token.
  563. func yaml_parser_scan(parser *yaml_parser_t, token *yaml_token_t) bool {
  564. // Erase the token object.
  565. *token = yaml_token_t{} // [Go] Is this necessary?
  566. // No tokens after STREAM-END or error.
  567. if parser.stream_end_produced || parser.error != yaml_NO_ERROR {
  568. return true
  569. }
  570. // Ensure that the tokens queue contains enough tokens.
  571. if !parser.token_available {
  572. if !yaml_parser_fetch_more_tokens(parser) {
  573. return false
  574. }
  575. }
  576. // Fetch the next token from the queue.
  577. *token = parser.tokens[parser.tokens_head]
  578. parser.tokens_head++
  579. parser.tokens_parsed++
  580. parser.token_available = false
  581. if token.typ == yaml_STREAM_END_TOKEN {
  582. parser.stream_end_produced = true
  583. }
  584. return true
  585. }
  586. // Set the scanner error and return false.
  587. func yaml_parser_set_scanner_error(parser *yaml_parser_t, context string, context_mark yaml_mark_t, problem string) bool {
  588. parser.error = yaml_SCANNER_ERROR
  589. parser.context = context
  590. parser.context_mark = context_mark
  591. parser.problem = problem
  592. parser.problem_mark = parser.mark
  593. return false
  594. }
  595. func yaml_parser_set_scanner_tag_error(parser *yaml_parser_t, directive bool, context_mark yaml_mark_t, problem string) bool {
  596. context := "while parsing a tag"
  597. if directive {
  598. context = "while parsing a %TAG directive"
  599. }
  600. return yaml_parser_set_scanner_error(parser, context, context_mark, problem)
  601. }
  602. func trace(args ...interface{}) func() {
  603. pargs := append([]interface{}{"+++"}, args...)
  604. fmt.Println(pargs...)
  605. pargs = append([]interface{}{"---"}, args...)
  606. return func() { fmt.Println(pargs...) }
  607. }
  608. // Ensure that the tokens queue contains at least one token which can be
  609. // returned to the Parser.
  610. func yaml_parser_fetch_more_tokens(parser *yaml_parser_t) bool {
  611. // While we need more tokens to fetch, do it.
  612. for {
  613. // Check if we really need to fetch more tokens.
  614. need_more_tokens := false
  615. // [Go] When parsing flow items, force the queue to have at least
  616. // two items so that comments after commas may be associated
  617. // with the value being parsed before them.
  618. if parser.tokens_head == len(parser.tokens) || parser.flow_level > 0 && parser.tokens_head >= len(parser.tokens)-1 {
  619. // Queue is empty or has just one element inside a flow context.
  620. need_more_tokens = true
  621. } else {
  622. // Check if any potential simple key may occupy the head position.
  623. if !yaml_parser_stale_simple_keys(parser) {
  624. return false
  625. }
  626. for i := range parser.simple_keys {
  627. simple_key := &parser.simple_keys[i]
  628. if simple_key.possible && simple_key.token_number == parser.tokens_parsed {
  629. need_more_tokens = true
  630. break
  631. }
  632. }
  633. }
  634. // We are finished.
  635. if !need_more_tokens {
  636. break
  637. }
  638. // Fetch the next token.
  639. if !yaml_parser_fetch_next_token(parser) {
  640. return false
  641. }
  642. }
  643. parser.token_available = true
  644. return true
  645. }
  646. // The dispatcher for token fetchers.
  647. func yaml_parser_fetch_next_token(parser *yaml_parser_t) (ok bool) {
  648. // Ensure that the buffer is initialized.
  649. if parser.unread < 1 && !yaml_parser_update_buffer(parser, 1) {
  650. return false
  651. }
  652. // Check if we just started scanning. Fetch STREAM-START then.
  653. if !parser.stream_start_produced {
  654. return yaml_parser_fetch_stream_start(parser)
  655. }
  656. // Eat whitespaces and comments until we reach the next token.
  657. if !yaml_parser_scan_to_next_token(parser) {
  658. return false
  659. }
  660. // Remove obsolete potential simple keys.
  661. if !yaml_parser_stale_simple_keys(parser) {
  662. return false
  663. }
  664. // Check the indentation level against the current column.
  665. if !yaml_parser_unroll_indent(parser, parser.mark.column) {
  666. return false
  667. }
  668. // Ensure that the buffer contains at least 4 characters. 4 is the length
  669. // of the longest indicators ('--- ' and '... ').
  670. if parser.unread < 4 && !yaml_parser_update_buffer(parser, 4) {
  671. return false
  672. }
  673. // Is it the end of the stream?
  674. if is_z(parser.buffer, parser.buffer_pos) {
  675. return yaml_parser_fetch_stream_end(parser)
  676. }
  677. // Is it a directive?
  678. if parser.mark.column == 0 && parser.buffer[parser.buffer_pos] == '%' {
  679. return yaml_parser_fetch_directive(parser)
  680. }
  681. buf := parser.buffer
  682. pos := parser.buffer_pos
  683. // Is it the document start indicator?
  684. if parser.mark.column == 0 && buf[pos] == '-' && buf[pos+1] == '-' && buf[pos+2] == '-' && is_blankz(buf, pos+3) {
  685. return yaml_parser_fetch_document_indicator(parser, yaml_DOCUMENT_START_TOKEN)
  686. }
  687. // Is it the document end indicator?
  688. if parser.mark.column == 0 && buf[pos] == '.' && buf[pos+1] == '.' && buf[pos+2] == '.' && is_blankz(buf, pos+3) {
  689. return yaml_parser_fetch_document_indicator(parser, yaml_DOCUMENT_END_TOKEN)
  690. }
  691. comment_mark := parser.mark
  692. if parser.flow_level > 0 && buf[pos] == ',' && len(parser.tokens) > 0 {
  693. // Associate any following comments with the prior token.
  694. comment_mark = parser.tokens[len(parser.tokens)-1].start_mark
  695. }
  696. defer func() {
  697. if !ok {
  698. return
  699. }
  700. if !yaml_parser_scan_line_comment(parser, comment_mark) {
  701. ok = false
  702. return
  703. }
  704. if !yaml_parser_scan_foot_comment(parser, comment_mark) {
  705. ok = false
  706. return
  707. }
  708. }()
  709. // Is it the flow sequence start indicator?
  710. if buf[pos] == '[' {
  711. return yaml_parser_fetch_flow_collection_start(parser, yaml_FLOW_SEQUENCE_START_TOKEN)
  712. }
  713. // Is it the flow mapping start indicator?
  714. if parser.buffer[parser.buffer_pos] == '{' {
  715. return yaml_parser_fetch_flow_collection_start(parser, yaml_FLOW_MAPPING_START_TOKEN)
  716. }
  717. // Is it the flow sequence end indicator?
  718. if parser.buffer[parser.buffer_pos] == ']' {
  719. return yaml_parser_fetch_flow_collection_end(parser,
  720. yaml_FLOW_SEQUENCE_END_TOKEN)
  721. }
  722. // Is it the flow mapping end indicator?
  723. if parser.buffer[parser.buffer_pos] == '}' {
  724. return yaml_parser_fetch_flow_collection_end(parser,
  725. yaml_FLOW_MAPPING_END_TOKEN)
  726. }
  727. // Is it the flow entry indicator?
  728. if parser.buffer[parser.buffer_pos] == ',' {
  729. return yaml_parser_fetch_flow_entry(parser)
  730. }
  731. // Is it the block entry indicator?
  732. if parser.buffer[parser.buffer_pos] == '-' && is_blankz(parser.buffer, parser.buffer_pos+1) {
  733. return yaml_parser_fetch_block_entry(parser)
  734. }
  735. // Is it the key indicator?
  736. if parser.buffer[parser.buffer_pos] == '?' && (parser.flow_level > 0 || is_blankz(parser.buffer, parser.buffer_pos+1)) {
  737. return yaml_parser_fetch_key(parser)
  738. }
  739. // Is it the value indicator?
  740. if parser.buffer[parser.buffer_pos] == ':' && (parser.flow_level > 0 || is_blankz(parser.buffer, parser.buffer_pos+1)) {
  741. return yaml_parser_fetch_value(parser)
  742. }
  743. // Is it an alias?
  744. if parser.buffer[parser.buffer_pos] == '*' {
  745. return yaml_parser_fetch_anchor(parser, yaml_ALIAS_TOKEN)
  746. }
  747. // Is it an anchor?
  748. if parser.buffer[parser.buffer_pos] == '&' {
  749. return yaml_parser_fetch_anchor(parser, yaml_ANCHOR_TOKEN)
  750. }
  751. // Is it a tag?
  752. if parser.buffer[parser.buffer_pos] == '!' {
  753. return yaml_parser_fetch_tag(parser)
  754. }
  755. // Is it a literal scalar?
  756. if parser.buffer[parser.buffer_pos] == '|' && parser.flow_level == 0 {
  757. return yaml_parser_fetch_block_scalar(parser, true)
  758. }
  759. // Is it a folded scalar?
  760. if parser.buffer[parser.buffer_pos] == '>' && parser.flow_level == 0 {
  761. return yaml_parser_fetch_block_scalar(parser, false)
  762. }
  763. // Is it a single-quoted scalar?
  764. if parser.buffer[parser.buffer_pos] == '\'' {
  765. return yaml_parser_fetch_flow_scalar(parser, true)
  766. }
  767. // Is it a double-quoted scalar?
  768. if parser.buffer[parser.buffer_pos] == '"' {
  769. return yaml_parser_fetch_flow_scalar(parser, false)
  770. }
  771. // Is it a plain scalar?
  772. //
  773. // A plain scalar may start with any non-blank characters except
  774. //
  775. // '-', '?', ':', ',', '[', ']', '{', '}',
  776. // '#', '&', '*', '!', '|', '>', '\'', '\"',
  777. // '%', '@', '`'.
  778. //
  779. // In the block context (and, for the '-' indicator, in the flow context
  780. // too), it may also start with the characters
  781. //
  782. // '-', '?', ':'
  783. //
  784. // if it is followed by a non-space character.
  785. //
  786. // The last rule is more restrictive than the specification requires.
  787. // [Go] TODO Make this logic more reasonable.
  788. //switch parser.buffer[parser.buffer_pos] {
  789. //case '-', '?', ':', ',', '?', '-', ',', ':', ']', '[', '}', '{', '&', '#', '!', '*', '>', '|', '"', '\'', '@', '%', '-', '`':
  790. //}
  791. if !(is_blankz(parser.buffer, parser.buffer_pos) || parser.buffer[parser.buffer_pos] == '-' ||
  792. parser.buffer[parser.buffer_pos] == '?' || parser.buffer[parser.buffer_pos] == ':' ||
  793. parser.buffer[parser.buffer_pos] == ',' || parser.buffer[parser.buffer_pos] == '[' ||
  794. parser.buffer[parser.buffer_pos] == ']' || parser.buffer[parser.buffer_pos] == '{' ||
  795. parser.buffer[parser.buffer_pos] == '}' || parser.buffer[parser.buffer_pos] == '#' ||
  796. parser.buffer[parser.buffer_pos] == '&' || parser.buffer[parser.buffer_pos] == '*' ||
  797. parser.buffer[parser.buffer_pos] == '!' || parser.buffer[parser.buffer_pos] == '|' ||
  798. parser.buffer[parser.buffer_pos] == '>' || parser.buffer[parser.buffer_pos] == '\'' ||
  799. parser.buffer[parser.buffer_pos] == '"' || parser.buffer[parser.buffer_pos] == '%' ||
  800. parser.buffer[parser.buffer_pos] == '@' || parser.buffer[parser.buffer_pos] == '`') ||
  801. (parser.buffer[parser.buffer_pos] == '-' && !is_blank(parser.buffer, parser.buffer_pos+1)) ||
  802. (parser.flow_level == 0 &&
  803. (parser.buffer[parser.buffer_pos] == '?' || parser.buffer[parser.buffer_pos] == ':') &&
  804. !is_blankz(parser.buffer, parser.buffer_pos+1)) {
  805. return yaml_parser_fetch_plain_scalar(parser)
  806. }
  807. // If we don't determine the token type so far, it is an error.
  808. return yaml_parser_set_scanner_error(parser,
  809. "while scanning for the next token", parser.mark,
  810. "found character that cannot start any token")
  811. }
  812. // Check the list of potential simple keys and remove the positions that
  813. // cannot contain simple keys anymore.
  814. func yaml_parser_stale_simple_keys(parser *yaml_parser_t) bool {
  815. // Check for a potential simple key for each flow level.
  816. for i := range parser.simple_keys {
  817. simple_key := &parser.simple_keys[i]
  818. // The specification requires that a simple key
  819. //
  820. // - is limited to a single line,
  821. // - is shorter than 1024 characters.
  822. if simple_key.possible && (simple_key.mark.line < parser.mark.line || simple_key.mark.index+1024 < parser.mark.index) {
  823. // Check if the potential simple key to be removed is required.
  824. if simple_key.required {
  825. return yaml_parser_set_scanner_error(parser,
  826. "while scanning a simple key", simple_key.mark,
  827. "could not find expected ':'")
  828. }
  829. simple_key.possible = false
  830. }
  831. }
  832. return true
  833. }
  834. // Check if a simple key may start at the current position and add it if
  835. // needed.
  836. func yaml_parser_save_simple_key(parser *yaml_parser_t) bool {
  837. // A simple key is required at the current position if the scanner is in
  838. // the block context and the current column coincides with the indentation
  839. // level.
  840. required := parser.flow_level == 0 && parser.indent == parser.mark.column
  841. //
  842. // If the current position may start a simple key, save it.
  843. //
  844. if parser.simple_key_allowed {
  845. simple_key := yaml_simple_key_t{
  846. possible: true,
  847. required: required,
  848. token_number: parser.tokens_parsed + (len(parser.tokens) - parser.tokens_head),
  849. }
  850. simple_key.mark = parser.mark
  851. if !yaml_parser_remove_simple_key(parser) {
  852. return false
  853. }
  854. parser.simple_keys[len(parser.simple_keys)-1] = simple_key
  855. }
  856. return true
  857. }
  858. // Remove a potential simple key at the current flow level.
  859. func yaml_parser_remove_simple_key(parser *yaml_parser_t) bool {
  860. i := len(parser.simple_keys) - 1
  861. if parser.simple_keys[i].possible {
  862. // If the key is required, it is an error.
  863. if parser.simple_keys[i].required {
  864. return yaml_parser_set_scanner_error(parser,
  865. "while scanning a simple key", parser.simple_keys[i].mark,
  866. "could not find expected ':'")
  867. }
  868. }
  869. // Remove the key from the stack.
  870. parser.simple_keys[i].possible = false
  871. return true
  872. }
  873. // Increase the flow level and resize the simple key list if needed.
  874. func yaml_parser_increase_flow_level(parser *yaml_parser_t) bool {
  875. // Reset the simple key on the next level.
  876. parser.simple_keys = append(parser.simple_keys, yaml_simple_key_t{})
  877. // Increase the flow level.
  878. parser.flow_level++
  879. return true
  880. }
  881. // Decrease the flow level.
  882. func yaml_parser_decrease_flow_level(parser *yaml_parser_t) bool {
  883. if parser.flow_level > 0 {
  884. parser.flow_level--
  885. parser.simple_keys = parser.simple_keys[:len(parser.simple_keys)-1]
  886. }
  887. return true
  888. }
  889. // Push the current indentation level to the stack and set the new level
  890. // the current column is greater than the indentation level. In this case,
  891. // append or insert the specified token into the token queue.
  892. func yaml_parser_roll_indent(parser *yaml_parser_t, column, number int, typ yaml_token_type_t, mark yaml_mark_t) bool {
  893. // In the flow context, do nothing.
  894. if parser.flow_level > 0 {
  895. return true
  896. }
  897. if parser.indent < column {
  898. // Push the current indentation level to the stack and set the new
  899. // indentation level.
  900. parser.indents = append(parser.indents, parser.indent)
  901. parser.indent = column
  902. // Create a token and insert it into the queue.
  903. token := yaml_token_t{
  904. typ: typ,
  905. start_mark: mark,
  906. end_mark: mark,
  907. }
  908. if number > -1 {
  909. number -= parser.tokens_parsed
  910. }
  911. yaml_insert_token(parser, number, &token)
  912. }
  913. return true
  914. }
  915. // Pop indentation levels from the indents stack until the current level
  916. // becomes less or equal to the column. For each indentation level, append
  917. // the BLOCK-END token.
  918. func yaml_parser_unroll_indent(parser *yaml_parser_t, column int) bool {
  919. // In the flow context, do nothing.
  920. if parser.flow_level > 0 {
  921. return true
  922. }
  923. // Loop through the indentation levels in the stack.
  924. for parser.indent > column {
  925. // Create a token and append it to the queue.
  926. token := yaml_token_t{
  927. typ: yaml_BLOCK_END_TOKEN,
  928. start_mark: parser.mark,
  929. end_mark: parser.mark,
  930. }
  931. yaml_insert_token(parser, -1, &token)
  932. // Pop the indentation level.
  933. parser.indent = parser.indents[len(parser.indents)-1]
  934. parser.indents = parser.indents[:len(parser.indents)-1]
  935. }
  936. return true
  937. }
  938. // Initialize the scanner and produce the STREAM-START token.
  939. func yaml_parser_fetch_stream_start(parser *yaml_parser_t) bool {
  940. // Set the initial indentation.
  941. parser.indent = -1
  942. // Initialize the simple key stack.
  943. parser.simple_keys = append(parser.simple_keys, yaml_simple_key_t{})
  944. // A simple key is allowed at the beginning of the stream.
  945. parser.simple_key_allowed = true
  946. // We have started.
  947. parser.stream_start_produced = true
  948. // Create the STREAM-START token and append it to the queue.
  949. token := yaml_token_t{
  950. typ: yaml_STREAM_START_TOKEN,
  951. start_mark: parser.mark,
  952. end_mark: parser.mark,
  953. encoding: parser.encoding,
  954. }
  955. yaml_insert_token(parser, -1, &token)
  956. return true
  957. }
  958. // Produce the STREAM-END token and shut down the scanner.
  959. func yaml_parser_fetch_stream_end(parser *yaml_parser_t) bool {
  960. // Force new line.
  961. if parser.mark.column != 0 {
  962. parser.mark.column = 0
  963. parser.mark.line++
  964. }
  965. // Reset the indentation level.
  966. if !yaml_parser_unroll_indent(parser, -1) {
  967. return false
  968. }
  969. // Reset simple keys.
  970. if !yaml_parser_remove_simple_key(parser) {
  971. return false
  972. }
  973. parser.simple_key_allowed = false
  974. // Create the STREAM-END token and append it to the queue.
  975. token := yaml_token_t{
  976. typ: yaml_STREAM_END_TOKEN,
  977. start_mark: parser.mark,
  978. end_mark: parser.mark,
  979. }
  980. yaml_insert_token(parser, -1, &token)
  981. return true
  982. }
  983. // Produce a VERSION-DIRECTIVE or TAG-DIRECTIVE token.
  984. func yaml_parser_fetch_directive(parser *yaml_parser_t) bool {
  985. // Reset the indentation level.
  986. if !yaml_parser_unroll_indent(parser, -1) {
  987. return false
  988. }
  989. // Reset simple keys.
  990. if !yaml_parser_remove_simple_key(parser) {
  991. return false
  992. }
  993. parser.simple_key_allowed = false
  994. // Create the YAML-DIRECTIVE or TAG-DIRECTIVE token.
  995. token := yaml_token_t{}
  996. if !yaml_parser_scan_directive(parser, &token) {
  997. return false
  998. }
  999. // Append the token to the queue.
  1000. yaml_insert_token(parser, -1, &token)
  1001. return true
  1002. }
  1003. // Produce the DOCUMENT-START or DOCUMENT-END token.
  1004. func yaml_parser_fetch_document_indicator(parser *yaml_parser_t, typ yaml_token_type_t) bool {
  1005. // Reset the indentation level.
  1006. if !yaml_parser_unroll_indent(parser, -1) {
  1007. return false
  1008. }
  1009. // Reset simple keys.
  1010. if !yaml_parser_remove_simple_key(parser) {
  1011. return false
  1012. }
  1013. parser.simple_key_allowed = false
  1014. // Consume the token.
  1015. start_mark := parser.mark
  1016. skip(parser)
  1017. skip(parser)
  1018. skip(parser)
  1019. end_mark := parser.mark
  1020. // Create the DOCUMENT-START or DOCUMENT-END token.
  1021. token := yaml_token_t{
  1022. typ: typ,
  1023. start_mark: start_mark,
  1024. end_mark: end_mark,
  1025. }
  1026. // Append the token to the queue.
  1027. yaml_insert_token(parser, -1, &token)
  1028. return true
  1029. }
  1030. // Produce the FLOW-SEQUENCE-START or FLOW-MAPPING-START token.
  1031. func yaml_parser_fetch_flow_collection_start(parser *yaml_parser_t, typ yaml_token_type_t) bool {
  1032. // The indicators '[' and '{' may start a simple key.
  1033. if !yaml_parser_save_simple_key(parser) {
  1034. return false
  1035. }
  1036. // Increase the flow level.
  1037. if !yaml_parser_increase_flow_level(parser) {
  1038. return false
  1039. }
  1040. // A simple key may follow the indicators '[' and '{'.
  1041. parser.simple_key_allowed = true
  1042. // Consume the token.
  1043. start_mark := parser.mark
  1044. skip(parser)
  1045. end_mark := parser.mark
  1046. // Create the FLOW-SEQUENCE-START of FLOW-MAPPING-START token.
  1047. token := yaml_token_t{
  1048. typ: typ,
  1049. start_mark: start_mark,
  1050. end_mark: end_mark,
  1051. }
  1052. // Append the token to the queue.
  1053. yaml_insert_token(parser, -1, &token)
  1054. return true
  1055. }
  1056. // Produce the FLOW-SEQUENCE-END or FLOW-MAPPING-END token.
  1057. func yaml_parser_fetch_flow_collection_end(parser *yaml_parser_t, typ yaml_token_type_t) bool {
  1058. // Reset any potential simple key on the current flow level.
  1059. if !yaml_parser_remove_simple_key(parser) {
  1060. return false
  1061. }
  1062. // Decrease the flow level.
  1063. if !yaml_parser_decrease_flow_level(parser) {
  1064. return false
  1065. }
  1066. // No simple keys after the indicators ']' and '}'.
  1067. parser.simple_key_allowed = false
  1068. // Consume the token.
  1069. start_mark := parser.mark
  1070. skip(parser)
  1071. end_mark := parser.mark
  1072. // Create the FLOW-SEQUENCE-END of FLOW-MAPPING-END token.
  1073. token := yaml_token_t{
  1074. typ: typ,
  1075. start_mark: start_mark,
  1076. end_mark: end_mark,
  1077. }
  1078. // Append the token to the queue.
  1079. yaml_insert_token(parser, -1, &token)
  1080. return true
  1081. }
  1082. // Produce the FLOW-ENTRY token.
  1083. func yaml_parser_fetch_flow_entry(parser *yaml_parser_t) bool {
  1084. // Reset any potential simple keys on the current flow level.
  1085. if !yaml_parser_remove_simple_key(parser) {
  1086. return false
  1087. }
  1088. // Simple keys are allowed after ','.
  1089. parser.simple_key_allowed = true
  1090. // Consume the token.
  1091. start_mark := parser.mark
  1092. skip(parser)
  1093. end_mark := parser.mark
  1094. // Create the FLOW-ENTRY token and append it to the queue.
  1095. token := yaml_token_t{
  1096. typ: yaml_FLOW_ENTRY_TOKEN,
  1097. start_mark: start_mark,
  1098. end_mark: end_mark,
  1099. }
  1100. yaml_insert_token(parser, -1, &token)
  1101. return true
  1102. }
  1103. // Produce the BLOCK-ENTRY token.
  1104. func yaml_parser_fetch_block_entry(parser *yaml_parser_t) bool {
  1105. // Check if the scanner is in the block context.
  1106. if parser.flow_level == 0 {
  1107. // Check if we are allowed to start a new entry.
  1108. if !parser.simple_key_allowed {
  1109. return yaml_parser_set_scanner_error(parser, "", parser.mark,
  1110. "block sequence entries are not allowed in this context")
  1111. }
  1112. // Add the BLOCK-SEQUENCE-START token if needed.
  1113. if !yaml_parser_roll_indent(parser, parser.mark.column, -1, yaml_BLOCK_SEQUENCE_START_TOKEN, parser.mark) {
  1114. return false
  1115. }
  1116. } else {
  1117. // It is an error for the '-' indicator to occur in the flow context,
  1118. // but we let the Parser detect and report about it because the Parser
  1119. // is able to point to the context.
  1120. }
  1121. // Reset any potential simple keys on the current flow level.
  1122. if !yaml_parser_remove_simple_key(parser) {
  1123. return false
  1124. }
  1125. // Simple keys are allowed after '-'.
  1126. parser.simple_key_allowed = true
  1127. // Consume the token.
  1128. start_mark := parser.mark
  1129. skip(parser)
  1130. end_mark := parser.mark
  1131. // Create the BLOCK-ENTRY token and append it to the queue.
  1132. token := yaml_token_t{
  1133. typ: yaml_BLOCK_ENTRY_TOKEN,
  1134. start_mark: start_mark,
  1135. end_mark: end_mark,
  1136. }
  1137. yaml_insert_token(parser, -1, &token)
  1138. return true
  1139. }
  1140. // Produce the KEY token.
  1141. func yaml_parser_fetch_key(parser *yaml_parser_t) bool {
  1142. // In the block context, additional checks are required.
  1143. if parser.flow_level == 0 {
  1144. // Check if we are allowed to start a new key (not nessesary simple).
  1145. if !parser.simple_key_allowed {
  1146. return yaml_parser_set_scanner_error(parser, "", parser.mark,
  1147. "mapping keys are not allowed in this context")
  1148. }
  1149. // Add the BLOCK-MAPPING-START token if needed.
  1150. if !yaml_parser_roll_indent(parser, parser.mark.column, -1, yaml_BLOCK_MAPPING_START_TOKEN, parser.mark) {
  1151. return false
  1152. }
  1153. }
  1154. // Reset any potential simple keys on the current flow level.
  1155. if !yaml_parser_remove_simple_key(parser) {
  1156. return false
  1157. }
  1158. // Simple keys are allowed after '?' in the block context.
  1159. parser.simple_key_allowed = parser.flow_level == 0
  1160. // Consume the token.
  1161. start_mark := parser.mark
  1162. skip(parser)
  1163. end_mark := parser.mark
  1164. // Create the KEY token and append it to the queue.
  1165. token := yaml_token_t{
  1166. typ: yaml_KEY_TOKEN,
  1167. start_mark: start_mark,
  1168. end_mark: end_mark,
  1169. }
  1170. yaml_insert_token(parser, -1, &token)
  1171. return true
  1172. }
  1173. // Produce the VALUE token.
  1174. func yaml_parser_fetch_value(parser *yaml_parser_t) bool {
  1175. simple_key := &parser.simple_keys[len(parser.simple_keys)-1]
  1176. // Have we found a simple key?
  1177. if simple_key.possible {
  1178. // Create the KEY token and insert it into the queue.
  1179. token := yaml_token_t{
  1180. typ: yaml_KEY_TOKEN,
  1181. start_mark: simple_key.mark,
  1182. end_mark: simple_key.mark,
  1183. }
  1184. yaml_insert_token(parser, simple_key.token_number-parser.tokens_parsed, &token)
  1185. // In the block context, we may need to add the BLOCK-MAPPING-START token.
  1186. if !yaml_parser_roll_indent(parser, simple_key.mark.column,
  1187. simple_key.token_number,
  1188. yaml_BLOCK_MAPPING_START_TOKEN, simple_key.mark) {
  1189. return false
  1190. }
  1191. // Remove the simple key.
  1192. simple_key.possible = false
  1193. // A simple key cannot follow another simple key.
  1194. parser.simple_key_allowed = false
  1195. } else {
  1196. // The ':' indicator follows a complex key.
  1197. // In the block context, extra checks are required.
  1198. if parser.flow_level == 0 {
  1199. // Check if we are allowed to start a complex value.
  1200. if !parser.simple_key_allowed {
  1201. return yaml_parser_set_scanner_error(parser, "", parser.mark,
  1202. "mapping values are not allowed in this context")
  1203. }
  1204. // Add the BLOCK-MAPPING-START token if needed.
  1205. if !yaml_parser_roll_indent(parser, parser.mark.column, -1, yaml_BLOCK_MAPPING_START_TOKEN, parser.mark) {
  1206. return false
  1207. }
  1208. }
  1209. // Simple keys after ':' are allowed in the block context.
  1210. parser.simple_key_allowed = parser.flow_level == 0
  1211. }
  1212. // Consume the token.
  1213. start_mark := parser.mark
  1214. skip(parser)
  1215. end_mark := parser.mark
  1216. // Create the VALUE token and append it to the queue.
  1217. token := yaml_token_t{
  1218. typ: yaml_VALUE_TOKEN,
  1219. start_mark: start_mark,
  1220. end_mark: end_mark,
  1221. }
  1222. yaml_insert_token(parser, -1, &token)
  1223. return true
  1224. }
  1225. // Produce the ALIAS or ANCHOR token.
  1226. func yaml_parser_fetch_anchor(parser *yaml_parser_t, typ yaml_token_type_t) bool {
  1227. // An anchor or an alias could be a simple key.
  1228. if !yaml_parser_save_simple_key(parser) {
  1229. return false
  1230. }
  1231. // A simple key cannot follow an anchor or an alias.
  1232. parser.simple_key_allowed = false
  1233. // Create the ALIAS or ANCHOR token and append it to the queue.
  1234. var token yaml_token_t
  1235. if !yaml_parser_scan_anchor(parser, &token, typ) {
  1236. return false
  1237. }
  1238. yaml_insert_token(parser, -1, &token)
  1239. return true
  1240. }
  1241. // Produce the TAG token.
  1242. func yaml_parser_fetch_tag(parser *yaml_parser_t) bool {
  1243. // A tag could be a simple key.
  1244. if !yaml_parser_save_simple_key(parser) {
  1245. return false
  1246. }
  1247. // A simple key cannot follow a tag.
  1248. parser.simple_key_allowed = false
  1249. // Create the TAG token and append it to the queue.
  1250. var token yaml_token_t
  1251. if !yaml_parser_scan_tag(parser, &token) {
  1252. return false
  1253. }
  1254. yaml_insert_token(parser, -1, &token)
  1255. return true
  1256. }
  1257. // Produce the SCALAR(...,literal) or SCALAR(...,folded) tokens.
  1258. func yaml_parser_fetch_block_scalar(parser *yaml_parser_t, literal bool) bool {
  1259. // Remove any potential simple keys.
  1260. if !yaml_parser_remove_simple_key(parser) {
  1261. return false
  1262. }
  1263. // A simple key may follow a block scalar.
  1264. parser.simple_key_allowed = true
  1265. // Create the SCALAR token and append it to the queue.
  1266. var token yaml_token_t
  1267. if !yaml_parser_scan_block_scalar(parser, &token, literal) {
  1268. return false
  1269. }
  1270. yaml_insert_token(parser, -1, &token)
  1271. return true
  1272. }
  1273. // Produce the SCALAR(...,single-quoted) or SCALAR(...,double-quoted) tokens.
  1274. func yaml_parser_fetch_flow_scalar(parser *yaml_parser_t, single bool) bool {
  1275. // A plain scalar could be a simple key.
  1276. if !yaml_parser_save_simple_key(parser) {
  1277. return false
  1278. }
  1279. // A simple key cannot follow a flow scalar.
  1280. parser.simple_key_allowed = false
  1281. // Create the SCALAR token and append it to the queue.
  1282. var token yaml_token_t
  1283. if !yaml_parser_scan_flow_scalar(parser, &token, single) {
  1284. return false
  1285. }
  1286. yaml_insert_token(parser, -1, &token)
  1287. return true
  1288. }
  1289. // Produce the SCALAR(...,plain) token.
  1290. func yaml_parser_fetch_plain_scalar(parser *yaml_parser_t) bool {
  1291. // A plain scalar could be a simple key.
  1292. if !yaml_parser_save_simple_key(parser) {
  1293. return false
  1294. }
  1295. // A simple key cannot follow a flow scalar.
  1296. parser.simple_key_allowed = false
  1297. // Create the SCALAR token and append it to the queue.
  1298. var token yaml_token_t
  1299. if !yaml_parser_scan_plain_scalar(parser, &token) {
  1300. return false
  1301. }
  1302. yaml_insert_token(parser, -1, &token)
  1303. return true
  1304. }
  1305. // Eat whitespaces and comments until the next token is found.
  1306. func yaml_parser_scan_to_next_token(parser *yaml_parser_t) bool {
  1307. // Until the next token is not found.
  1308. for {
  1309. // Allow the BOM mark to start a line.
  1310. if parser.unread < 1 && !yaml_parser_update_buffer(parser, 1) {
  1311. return false
  1312. }
  1313. if parser.mark.column == 0 && is_bom(parser.buffer, parser.buffer_pos) {
  1314. skip(parser)
  1315. }
  1316. // Eat whitespaces.
  1317. // Tabs are allowed:
  1318. // - in the flow context
  1319. // - in the block context, but not at the beginning of the line or
  1320. // after '-', '?', or ':' (complex value).
  1321. if parser.unread < 1 && !yaml_parser_update_buffer(parser, 1) {
  1322. return false
  1323. }
  1324. for parser.buffer[parser.buffer_pos] == ' ' || ((parser.flow_level > 0 || !parser.simple_key_allowed) && parser.buffer[parser.buffer_pos] == '\t') {
  1325. skip(parser)
  1326. if parser.unread < 1 && !yaml_parser_update_buffer(parser, 1) {
  1327. return false
  1328. }
  1329. }
  1330. // Eat a comment until a line break.
  1331. if parser.buffer[parser.buffer_pos] == '#' {
  1332. if !yaml_parser_scan_head_comment(parser, parser.mark) {
  1333. return false
  1334. }
  1335. }
  1336. // If it is a line break, eat it.
  1337. if is_break(parser.buffer, parser.buffer_pos) {
  1338. if parser.unread < 2 && !yaml_parser_update_buffer(parser, 2) {
  1339. return false
  1340. }
  1341. skip_line(parser)
  1342. // In the block context, a new line may start a simple key.
  1343. if parser.flow_level == 0 {
  1344. parser.simple_key_allowed = true
  1345. }
  1346. } else {
  1347. break // We have found a token.
  1348. }
  1349. }
  1350. return true
  1351. }
  1352. // Scan a YAML-DIRECTIVE or TAG-DIRECTIVE token.
  1353. //
  1354. // Scope:
  1355. // %YAML 1.1 # a comment \n
  1356. // ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  1357. // %TAG !yaml! tag:yaml.org,2002: \n
  1358. // ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  1359. //
  1360. func yaml_parser_scan_directive(parser *yaml_parser_t, token *yaml_token_t) bool {
  1361. // Eat '%'.
  1362. start_mark := parser.mark
  1363. skip(parser)
  1364. // Scan the directive name.
  1365. var name []byte
  1366. if !yaml_parser_scan_directive_name(parser, start_mark, &name) {
  1367. return false
  1368. }
  1369. // Is it a YAML directive?
  1370. if bytes.Equal(name, []byte("YAML")) {
  1371. // Scan the VERSION directive value.
  1372. var major, minor int8
  1373. if !yaml_parser_scan_version_directive_value(parser, start_mark, &major, &minor) {
  1374. return false
  1375. }
  1376. end_mark := parser.mark
  1377. // Create a VERSION-DIRECTIVE token.
  1378. *token = yaml_token_t{
  1379. typ: yaml_VERSION_DIRECTIVE_TOKEN,
  1380. start_mark: start_mark,
  1381. end_mark: end_mark,
  1382. major: major,
  1383. minor: minor,
  1384. }
  1385. // Is it a TAG directive?
  1386. } else if bytes.Equal(name, []byte("TAG")) {
  1387. // Scan the TAG directive value.
  1388. var handle, prefix []byte
  1389. if !yaml_parser_scan_tag_directive_value(parser, start_mark, &handle, &prefix) {
  1390. return false
  1391. }
  1392. end_mark := parser.mark
  1393. // Create a TAG-DIRECTIVE token.
  1394. *token = yaml_token_t{
  1395. typ: yaml_TAG_DIRECTIVE_TOKEN,
  1396. start_mark: start_mark,
  1397. end_mark: end_mark,
  1398. value: handle,
  1399. prefix: prefix,
  1400. }
  1401. // Unknown directive.
  1402. } else {
  1403. yaml_parser_set_scanner_error(parser, "while scanning a directive",
  1404. start_mark, "found unknown directive name")
  1405. return false
  1406. }
  1407. // Eat the rest of the line including any comments.
  1408. if parser.unread < 1 && !yaml_parser_update_buffer(parser, 1) {
  1409. return false
  1410. }
  1411. for is_blank(parser.buffer, parser.buffer_pos) {
  1412. skip(parser)
  1413. if parser.unread < 1 && !yaml_parser_update_buffer(parser, 1) {
  1414. return false
  1415. }
  1416. }
  1417. if parser.buffer[parser.buffer_pos] == '#' {
  1418. // [Go] Discard this inline comment for the time being.
  1419. //if !yaml_parser_scan_line_comment(parser, start_mark) {
  1420. // return false
  1421. //}
  1422. for !is_breakz(parser.buffer, parser.buffer_pos) {
  1423. skip(parser)
  1424. if parser.unread < 1 && !yaml_parser_update_buffer(parser, 1) {
  1425. return false
  1426. }
  1427. }
  1428. }
  1429. // Check if we are at the end of the line.
  1430. if !is_breakz(parser.buffer, parser.buffer_pos) {
  1431. yaml_parser_set_scanner_error(parser, "while scanning a directive",
  1432. start_mark, "did not find expected comment or line break")
  1433. return false
  1434. }
  1435. // Eat a line break.
  1436. if is_break(parser.buffer, parser.buffer_pos) {
  1437. if parser.unread < 2 && !yaml_parser_update_buffer(parser, 2) {
  1438. return false
  1439. }
  1440. skip_line(parser)
  1441. }
  1442. return true
  1443. }
  1444. // Scan the directive name.
  1445. //
  1446. // Scope:
  1447. // %YAML 1.1 # a comment \n
  1448. // ^^^^
  1449. // %TAG !yaml! tag:yaml.org,2002: \n
  1450. // ^^^
  1451. //
  1452. func yaml_parser_scan_directive_name(parser *yaml_parser_t, start_mark yaml_mark_t, name *[]byte) bool {
  1453. // Consume the directive name.
  1454. if parser.unread < 1 && !yaml_parser_update_buffer(parser, 1) {
  1455. return false
  1456. }
  1457. var s []byte
  1458. for is_alpha(parser.buffer, parser.buffer_pos) {
  1459. s = read(parser, s)
  1460. if parser.unread < 1 && !yaml_parser_update_buffer(parser, 1) {
  1461. return false
  1462. }
  1463. }
  1464. // Check if the name is empty.
  1465. if len(s) == 0 {
  1466. yaml_parser_set_scanner_error(parser, "while scanning a directive",
  1467. start_mark, "could not find expected directive name")
  1468. return false
  1469. }
  1470. // Check for an blank character after the name.
  1471. if !is_blankz(parser.buffer, parser.buffer_pos) {
  1472. yaml_parser_set_scanner_error(parser, "while scanning a directive",
  1473. start_mark, "found unexpected non-alphabetical character")
  1474. return false
  1475. }
  1476. *name = s
  1477. return true
  1478. }
  1479. // Scan the value of VERSION-DIRECTIVE.
  1480. //
  1481. // Scope:
  1482. // %YAML 1.1 # a comment \n
  1483. // ^^^^^^
  1484. func yaml_parser_scan_version_directive_value(parser *yaml_parser_t, start_mark yaml_mark_t, major, minor *int8) bool {
  1485. // Eat whitespaces.
  1486. if parser.unread < 1 && !yaml_parser_update_buffer(parser, 1) {
  1487. return false
  1488. }
  1489. for is_blank(parser.buffer, parser.buffer_pos) {
  1490. skip(parser)
  1491. if parser.unread < 1 && !yaml_parser_update_buffer(parser, 1) {
  1492. return false
  1493. }
  1494. }
  1495. // Consume the major version number.
  1496. if !yaml_parser_scan_version_directive_number(parser, start_mark, major) {
  1497. return false
  1498. }
  1499. // Eat '.'.
  1500. if parser.buffer[parser.buffer_pos] != '.' {
  1501. return yaml_parser_set_scanner_error(parser, "while scanning a %YAML directive",
  1502. start_mark, "did not find expected digit or '.' character")
  1503. }
  1504. skip(parser)
  1505. // Consume the minor version number.
  1506. if !yaml_parser_scan_version_directive_number(parser, start_mark, minor) {
  1507. return false
  1508. }
  1509. return true
  1510. }
  1511. const max_number_length = 2
  1512. // Scan the version number of VERSION-DIRECTIVE.
  1513. //
  1514. // Scope:
  1515. // %YAML 1.1 # a comment \n
  1516. // ^
  1517. // %YAML 1.1 # a comment \n
  1518. // ^
  1519. func yaml_parser_scan_version_directive_number(parser *yaml_parser_t, start_mark yaml_mark_t, number *int8) bool {
  1520. // Repeat while the next character is digit.
  1521. if parser.unread < 1 && !yaml_parser_update_buffer(parser, 1) {
  1522. return false
  1523. }
  1524. var value, length int8
  1525. for is_digit(parser.buffer, parser.buffer_pos) {
  1526. // Check if the number is too long.
  1527. length++
  1528. if length > max_number_length {
  1529. return yaml_parser_set_scanner_error(parser, "while scanning a %YAML directive",
  1530. start_mark, "found extremely long version number")
  1531. }
  1532. value = value*10 + int8(as_digit(parser.buffer, parser.buffer_pos))
  1533. skip(parser)
  1534. if parser.unread < 1 && !yaml_parser_update_buffer(parser, 1) {
  1535. return false
  1536. }
  1537. }
  1538. // Check if the number was present.
  1539. if length == 0 {
  1540. return yaml_parser_set_scanner_error(parser, "while scanning a %YAML directive",
  1541. start_mark, "did not find expected version number")
  1542. }
  1543. *number = value
  1544. return true
  1545. }
  1546. // Scan the value of a TAG-DIRECTIVE token.
  1547. //
  1548. // Scope:
  1549. // %TAG !yaml! tag:yaml.org,2002: \n
  1550. // ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  1551. //
  1552. func yaml_parser_scan_tag_directive_value(parser *yaml_parser_t, start_mark yaml_mark_t, handle, prefix *[]byte) bool {
  1553. var handle_value, prefix_value []byte
  1554. // Eat whitespaces.
  1555. if parser.unread < 1 && !yaml_parser_update_buffer(parser, 1) {
  1556. return false
  1557. }
  1558. for is_blank(parser.buffer, parser.buffer_pos) {
  1559. skip(parser)
  1560. if parser.unread < 1 && !yaml_parser_update_buffer(parser, 1) {
  1561. return false
  1562. }
  1563. }
  1564. // Scan a handle.
  1565. if !yaml_parser_scan_tag_handle(parser, true, start_mark, &handle_value) {
  1566. return false
  1567. }
  1568. // Expect a whitespace.
  1569. if parser.unread < 1 && !yaml_parser_update_buffer(parser, 1) {
  1570. return false
  1571. }
  1572. if !is_blank(parser.buffer, parser.buffer_pos) {
  1573. yaml_parser_set_scanner_error(parser, "while scanning a %TAG directive",
  1574. start_mark, "did not find expected whitespace")
  1575. return false
  1576. }
  1577. // Eat whitespaces.
  1578. for is_blank(parser.buffer, parser.buffer_pos) {
  1579. skip(parser)
  1580. if parser.unread < 1 && !yaml_parser_update_buffer(parser, 1) {
  1581. return false
  1582. }
  1583. }
  1584. // Scan a prefix.
  1585. if !yaml_parser_scan_tag_uri(parser, true, nil, start_mark, &prefix_value) {
  1586. return false
  1587. }
  1588. // Expect a whitespace or line break.
  1589. if parser.unread < 1 && !yaml_parser_update_buffer(parser, 1) {
  1590. return false
  1591. }
  1592. if !is_blankz(parser.buffer, parser.buffer_pos) {
  1593. yaml_parser_set_scanner_error(parser, "while scanning a %TAG directive",
  1594. start_mark, "did not find expected whitespace or line break")
  1595. return false
  1596. }
  1597. *handle = handle_value
  1598. *prefix = prefix_value
  1599. return true
  1600. }
  1601. func yaml_parser_scan_anchor(parser *yaml_parser_t, token *yaml_token_t, typ yaml_token_type_t) bool {
  1602. var s []byte
  1603. // Eat the indicator character.
  1604. start_mark := parser.mark
  1605. skip(parser)
  1606. // Consume the value.
  1607. if parser.unread < 1 && !yaml_parser_update_buffer(parser, 1) {
  1608. return false
  1609. }
  1610. for is_alpha(parser.buffer, parser.buffer_pos) {
  1611. s = read(parser, s)
  1612. if parser.unread < 1 && !yaml_parser_update_buffer(parser, 1) {
  1613. return false
  1614. }
  1615. }
  1616. end_mark := parser.mark
  1617. /*
  1618. * Check if length of the anchor is greater than 0 and it is followed by
  1619. * a whitespace character or one of the indicators:
  1620. *
  1621. * '?', ':', ',', ']', '}', '%', '@', '`'.
  1622. */
  1623. if len(s) == 0 ||
  1624. !(is_blankz(parser.buffer, parser.buffer_pos) || parser.buffer[parser.buffer_pos] == '?' ||
  1625. parser.buffer[parser.buffer_pos] == ':' || parser.buffer[parser.buffer_pos] == ',' ||
  1626. parser.buffer[parser.buffer_pos] == ']' || parser.buffer[parser.buffer_pos] == '}' ||
  1627. parser.buffer[parser.buffer_pos] == '%' || parser.buffer[parser.buffer_pos] == '@' ||
  1628. parser.buffer[parser.buffer_pos] == '`') {
  1629. context := "while scanning an alias"
  1630. if typ == yaml_ANCHOR_TOKEN {
  1631. context = "while scanning an anchor"
  1632. }
  1633. yaml_parser_set_scanner_error(parser, context, start_mark,
  1634. "did not find expected alphabetic or numeric character")
  1635. return false
  1636. }
  1637. // Create a token.
  1638. *token = yaml_token_t{
  1639. typ: typ,
  1640. start_mark: start_mark,
  1641. end_mark: end_mark,
  1642. value: s,
  1643. }
  1644. return true
  1645. }
  1646. /*
  1647. * Scan a TAG token.
  1648. */
  1649. func yaml_parser_scan_tag(parser *yaml_parser_t, token *yaml_token_t) bool {
  1650. var handle, suffix []byte
  1651. start_mark := parser.mark
  1652. // Check if the tag is in the canonical form.
  1653. if parser.unread < 2 && !yaml_parser_update_buffer(parser, 2) {
  1654. return false
  1655. }
  1656. if parser.buffer[parser.buffer_pos+1] == '<' {
  1657. // Keep the handle as ''
  1658. // Eat '!<'
  1659. skip(parser)
  1660. skip(parser)
  1661. // Consume the tag value.
  1662. if !yaml_parser_scan_tag_uri(parser, false, nil, start_mark, &suffix) {
  1663. return false
  1664. }
  1665. // Check for '>' and eat it.
  1666. if parser.buffer[parser.buffer_pos] != '>' {
  1667. yaml_parser_set_scanner_error(parser, "while scanning a tag",
  1668. start_mark, "did not find the expected '>'")
  1669. return false
  1670. }
  1671. skip(parser)
  1672. } else {
  1673. // The tag has either the '!suffix' or the '!handle!suffix' form.
  1674. // First, try to scan a handle.
  1675. if !yaml_parser_scan_tag_handle(parser, false, start_mark, &handle) {
  1676. return false
  1677. }
  1678. // Check if it is, indeed, handle.
  1679. if handle[0] == '!' && len(handle) > 1 && handle[len(handle)-1] == '!' {
  1680. // Scan the suffix now.
  1681. if !yaml_parser_scan_tag_uri(parser, false, nil, start_mark, &suffix) {
  1682. return false
  1683. }
  1684. } else {
  1685. // It wasn't a handle after all. Scan the rest of the tag.
  1686. if !yaml_parser_scan_tag_uri(parser, false, handle, start_mark, &suffix) {
  1687. return false
  1688. }
  1689. // Set the handle to '!'.
  1690. handle = []byte{'!'}
  1691. // A special case: the '!' tag. Set the handle to '' and the
  1692. // suffix to '!'.
  1693. if len(suffix) == 0 {
  1694. handle, suffix = suffix, handle
  1695. }
  1696. }
  1697. }
  1698. // Check the character which ends the tag.
  1699. if parser.unread < 1 && !yaml_parser_update_buffer(parser, 1) {
  1700. return false
  1701. }
  1702. if !is_blankz(parser.buffer, parser.buffer_pos) {
  1703. yaml_parser_set_scanner_error(parser, "while scanning a tag",
  1704. start_mark, "did not find expected whitespace or line break")
  1705. return false
  1706. }
  1707. end_mark := parser.mark
  1708. // Create a token.
  1709. *token = yaml_token_t{
  1710. typ: yaml_TAG_TOKEN,
  1711. start_mark: start_mark,
  1712. end_mark: end_mark,
  1713. value: handle,
  1714. suffix: suffix,
  1715. }
  1716. return true
  1717. }
  1718. // Scan a tag handle.
  1719. func yaml_parser_scan_tag_handle(parser *yaml_parser_t, directive bool, start_mark yaml_mark_t, handle *[]byte) bool {
  1720. // Check the initial '!' character.
  1721. if parser.unread < 1 && !yaml_parser_update_buffer(parser, 1) {
  1722. return false
  1723. }
  1724. if parser.buffer[parser.buffer_pos] != '!' {
  1725. yaml_parser_set_scanner_tag_error(parser, directive,
  1726. start_mark, "did not find expected '!'")
  1727. return false
  1728. }
  1729. var s []byte
  1730. // Copy the '!' character.
  1731. s = read(parser, s)
  1732. // Copy all subsequent alphabetical and numerical characters.
  1733. if parser.unread < 1 && !yaml_parser_update_buffer(parser, 1) {
  1734. return false
  1735. }
  1736. for is_alpha(parser.buffer, parser.buffer_pos) {
  1737. s = read(parser, s)
  1738. if parser.unread < 1 && !yaml_parser_update_buffer(parser, 1) {
  1739. return false
  1740. }
  1741. }
  1742. // Check if the trailing character is '!' and copy it.
  1743. if parser.buffer[parser.buffer_pos] == '!' {
  1744. s = read(parser, s)
  1745. } else {
  1746. // It's either the '!' tag or not really a tag handle. If it's a %TAG
  1747. // directive, it's an error. If it's a tag token, it must be a part of URI.
  1748. if directive && string(s) != "!" {
  1749. yaml_parser_set_scanner_tag_error(parser, directive,
  1750. start_mark, "did not find expected '!'")
  1751. return false
  1752. }
  1753. }
  1754. *handle = s
  1755. return true
  1756. }
  1757. // Scan a tag.
  1758. func yaml_parser_scan_tag_uri(parser *yaml_parser_t, directive bool, head []byte, start_mark yaml_mark_t, uri *[]byte) bool {
  1759. //size_t length = head ? strlen((char *)head) : 0
  1760. var s []byte
  1761. hasTag := len(head) > 0
  1762. // Copy the head if needed.
  1763. //
  1764. // Note that we don't copy the leading '!' character.
  1765. if len(head) > 1 {
  1766. s = append(s, head[1:]...)
  1767. }
  1768. // Scan the tag.
  1769. if parser.unread < 1 && !yaml_parser_update_buffer(parser, 1) {
  1770. return false
  1771. }
  1772. // The set of characters that may appear in URI is as follows:
  1773. //
  1774. // '0'-'9', 'A'-'Z', 'a'-'z', '_', '-', ';', '/', '?', ':', '@', '&',
  1775. // '=', '+', '$', ',', '.', '!', '~', '*', '\'', '(', ')', '[', ']',
  1776. // '%'.
  1777. // [Go] TODO Convert this into more reasonable logic.
  1778. for is_alpha(parser.buffer, parser.buffer_pos) || parser.buffer[parser.buffer_pos] == ';' ||
  1779. parser.buffer[parser.buffer_pos] == '/' || parser.buffer[parser.buffer_pos] == '?' ||
  1780. parser.buffer[parser.buffer_pos] == ':' || parser.buffer[parser.buffer_pos] == '@' ||
  1781. parser.buffer[parser.buffer_pos] == '&' || parser.buffer[parser.buffer_pos] == '=' ||
  1782. parser.buffer[parser.buffer_pos] == '+' || parser.buffer[parser.buffer_pos] == '$' ||
  1783. parser.buffer[parser.buffer_pos] == ',' || parser.buffer[parser.buffer_pos] == '.' ||
  1784. parser.buffer[parser.buffer_pos] == '!' || parser.buffer[parser.buffer_pos] == '~' ||
  1785. parser.buffer[parser.buffer_pos] == '*' || parser.buffer[parser.buffer_pos] == '\'' ||
  1786. parser.buffer[parser.buffer_pos] == '(' || parser.buffer[parser.buffer_pos] == ')' ||
  1787. parser.buffer[parser.buffer_pos] == '[' || parser.buffer[parser.buffer_pos] == ']' ||
  1788. parser.buffer[parser.buffer_pos] == '%' {
  1789. // Check if it is a URI-escape sequence.
  1790. if parser.buffer[parser.buffer_pos] == '%' {
  1791. if !yaml_parser_scan_uri_escapes(parser, directive, start_mark, &s) {
  1792. return false
  1793. }
  1794. } else {
  1795. s = read(parser, s)
  1796. }
  1797. if parser.unread < 1 && !yaml_parser_update_buffer(parser, 1) {
  1798. return false
  1799. }
  1800. hasTag = true
  1801. }
  1802. if !hasTag {
  1803. yaml_parser_set_scanner_tag_error(parser, directive,
  1804. start_mark, "did not find expected tag URI")
  1805. return false
  1806. }
  1807. *uri = s
  1808. return true
  1809. }
  1810. // Decode an URI-escape sequence corresponding to a single UTF-8 character.
  1811. func yaml_parser_scan_uri_escapes(parser *yaml_parser_t, directive bool, start_mark yaml_mark_t, s *[]byte) bool {
  1812. // Decode the required number of characters.
  1813. w := 1024
  1814. for w > 0 {
  1815. // Check for a URI-escaped octet.
  1816. if parser.unread < 3 && !yaml_parser_update_buffer(parser, 3) {
  1817. return false
  1818. }
  1819. if !(parser.buffer[parser.buffer_pos] == '%' &&
  1820. is_hex(parser.buffer, parser.buffer_pos+1) &&
  1821. is_hex(parser.buffer, parser.buffer_pos+2)) {
  1822. return yaml_parser_set_scanner_tag_error(parser, directive,
  1823. start_mark, "did not find URI escaped octet")
  1824. }
  1825. // Get the octet.
  1826. octet := byte((as_hex(parser.buffer, parser.buffer_pos+1) << 4) + as_hex(parser.buffer, parser.buffer_pos+2))
  1827. // If it is the leading octet, determine the length of the UTF-8 sequence.
  1828. if w == 1024 {
  1829. w = width(octet)
  1830. if w == 0 {
  1831. return yaml_parser_set_scanner_tag_error(parser, directive,
  1832. start_mark, "found an incorrect leading UTF-8 octet")
  1833. }
  1834. } else {
  1835. // Check if the trailing octet is correct.
  1836. if octet&0xC0 != 0x80 {
  1837. return yaml_parser_set_scanner_tag_error(parser, directive,
  1838. start_mark, "found an incorrect trailing UTF-8 octet")
  1839. }
  1840. }
  1841. // Copy the octet and move the pointers.
  1842. *s = append(*s, octet)
  1843. skip(parser)
  1844. skip(parser)
  1845. skip(parser)
  1846. w--
  1847. }
  1848. return true
  1849. }
  1850. // Scan a block scalar.
  1851. func yaml_parser_scan_block_scalar(parser *yaml_parser_t, token *yaml_token_t, literal bool) bool {
  1852. // Eat the indicator '|' or '>'.
  1853. start_mark := parser.mark
  1854. skip(parser)
  1855. // Scan the additional block scalar indicators.
  1856. if parser.unread < 1 && !yaml_parser_update_buffer(parser, 1) {
  1857. return false
  1858. }
  1859. // Check for a chomping indicator.
  1860. var chomping, increment int
  1861. if parser.buffer[parser.buffer_pos] == '+' || parser.buffer[parser.buffer_pos] == '-' {
  1862. // Set the chomping method and eat the indicator.
  1863. if parser.buffer[parser.buffer_pos] == '+' {
  1864. chomping = +1
  1865. } else {
  1866. chomping = -1
  1867. }
  1868. skip(parser)
  1869. // Check for an indentation indicator.
  1870. if parser.unread < 1 && !yaml_parser_update_buffer(parser, 1) {
  1871. return false
  1872. }
  1873. if is_digit(parser.buffer, parser.buffer_pos) {
  1874. // Check that the indentation is greater than 0.
  1875. if parser.buffer[parser.buffer_pos] == '0' {
  1876. yaml_parser_set_scanner_error(parser, "while scanning a block scalar",
  1877. start_mark, "found an indentation indicator equal to 0")
  1878. return false
  1879. }
  1880. // Get the indentation level and eat the indicator.
  1881. increment = as_digit(parser.buffer, parser.buffer_pos)
  1882. skip(parser)
  1883. }
  1884. } else if is_digit(parser.buffer, parser.buffer_pos) {
  1885. // Do the same as above, but in the opposite order.
  1886. if parser.buffer[parser.buffer_pos] == '0' {
  1887. yaml_parser_set_scanner_error(parser, "while scanning a block scalar",
  1888. start_mark, "found an indentation indicator equal to 0")
  1889. return false
  1890. }
  1891. increment = as_digit(parser.buffer, parser.buffer_pos)
  1892. skip(parser)
  1893. if parser.unread < 1 && !yaml_parser_update_buffer(parser, 1) {
  1894. return false
  1895. }
  1896. if parser.buffer[parser.buffer_pos] == '+' || parser.buffer[parser.buffer_pos] == '-' {
  1897. if parser.buffer[parser.buffer_pos] == '+' {
  1898. chomping = +1
  1899. } else {
  1900. chomping = -1
  1901. }
  1902. skip(parser)
  1903. }
  1904. }
  1905. // Eat whitespaces and comments to the end of the line.
  1906. if parser.unread < 1 && !yaml_parser_update_buffer(parser, 1) {
  1907. return false
  1908. }
  1909. for is_blank(parser.buffer, parser.buffer_pos) {
  1910. skip(parser)
  1911. if parser.unread < 1 && !yaml_parser_update_buffer(parser, 1) {
  1912. return false
  1913. }
  1914. }
  1915. if parser.buffer[parser.buffer_pos] == '#' {
  1916. if !yaml_parser_scan_line_comment(parser, start_mark) {
  1917. return false
  1918. }
  1919. }
  1920. // Check if we are at the end of the line.
  1921. if !is_breakz(parser.buffer, parser.buffer_pos) {
  1922. yaml_parser_set_scanner_error(parser, "while scanning a block scalar",
  1923. start_mark, "did not find expected comment or line break")
  1924. return false
  1925. }
  1926. // Eat a line break.
  1927. if is_break(parser.buffer, parser.buffer_pos) {
  1928. if parser.unread < 2 && !yaml_parser_update_buffer(parser, 2) {
  1929. return false
  1930. }
  1931. skip_line(parser)
  1932. }
  1933. end_mark := parser.mark
  1934. // Set the indentation level if it was specified.
  1935. var indent int
  1936. if increment > 0 {
  1937. if parser.indent >= 0 {
  1938. indent = parser.indent + increment
  1939. } else {
  1940. indent = increment
  1941. }
  1942. }
  1943. // Scan the leading line breaks and determine the indentation level if needed.
  1944. var s, leading_break, trailing_breaks []byte
  1945. if !yaml_parser_scan_block_scalar_breaks(parser, &indent, &trailing_breaks, start_mark, &end_mark) {
  1946. return false
  1947. }
  1948. // Scan the block scalar content.
  1949. if parser.unread < 1 && !yaml_parser_update_buffer(parser, 1) {
  1950. return false
  1951. }
  1952. var leading_blank, trailing_blank bool
  1953. for parser.mark.column == indent && !is_z(parser.buffer, parser.buffer_pos) {
  1954. // We are at the beginning of a non-empty line.
  1955. // Is it a trailing whitespace?
  1956. trailing_blank = is_blank(parser.buffer, parser.buffer_pos)
  1957. // Check if we need to fold the leading line break.
  1958. if !literal && !leading_blank && !trailing_blank && len(leading_break) > 0 && leading_break[0] == '\n' {
  1959. // Do we need to join the lines by space?
  1960. if len(trailing_breaks) == 0 {
  1961. s = append(s, ' ')
  1962. }
  1963. } else {
  1964. s = append(s, leading_break...)
  1965. }
  1966. leading_break = leading_break[:0]
  1967. // Append the remaining line breaks.
  1968. s = append(s, trailing_breaks...)
  1969. trailing_breaks = trailing_breaks[:0]
  1970. // Is it a leading whitespace?
  1971. leading_blank = is_blank(parser.buffer, parser.buffer_pos)
  1972. // Consume the current line.
  1973. for !is_breakz(parser.buffer, parser.buffer_pos) {
  1974. s = read(parser, s)
  1975. if parser.unread < 1 && !yaml_parser_update_buffer(parser, 1) {
  1976. return false
  1977. }
  1978. }
  1979. // Consume the line break.
  1980. if parser.unread < 2 && !yaml_parser_update_buffer(parser, 2) {
  1981. return false
  1982. }
  1983. leading_break = read_line(parser, leading_break)
  1984. // Eat the following indentation spaces and line breaks.
  1985. if !yaml_parser_scan_block_scalar_breaks(parser, &indent, &trailing_breaks, start_mark, &end_mark) {
  1986. return false
  1987. }
  1988. }
  1989. // Chomp the tail.
  1990. if chomping != -1 {
  1991. s = append(s, leading_break...)
  1992. }
  1993. if chomping == 1 {
  1994. s = append(s, trailing_breaks...)
  1995. }
  1996. // Create a token.
  1997. *token = yaml_token_t{
  1998. typ: yaml_SCALAR_TOKEN,
  1999. start_mark: start_mark,
  2000. end_mark: end_mark,
  2001. value: s,
  2002. style: yaml_LITERAL_SCALAR_STYLE,
  2003. }
  2004. if !literal {
  2005. token.style = yaml_FOLDED_SCALAR_STYLE
  2006. }
  2007. return true
  2008. }
  2009. // Scan indentation spaces and line breaks for a block scalar. Determine the
  2010. // indentation level if needed.
  2011. func yaml_parser_scan_block_scalar_breaks(parser *yaml_parser_t, indent *int, breaks *[]byte, start_mark yaml_mark_t, end_mark *yaml_mark_t) bool {
  2012. *end_mark = parser.mark
  2013. // Eat the indentation spaces and line breaks.
  2014. max_indent := 0
  2015. for {
  2016. // Eat the indentation spaces.
  2017. if parser.unread < 1 && !yaml_parser_update_buffer(parser, 1) {
  2018. return false
  2019. }
  2020. for (*indent == 0 || parser.mark.column < *indent) && is_space(parser.buffer, parser.buffer_pos) {
  2021. skip(parser)
  2022. if parser.unread < 1 && !yaml_parser_update_buffer(parser, 1) {
  2023. return false
  2024. }
  2025. }
  2026. if parser.mark.column > max_indent {
  2027. max_indent = parser.mark.column
  2028. }
  2029. // Check for a tab character messing the indentation.
  2030. if (*indent == 0 || parser.mark.column < *indent) && is_tab(parser.buffer, parser.buffer_pos) {
  2031. return yaml_parser_set_scanner_error(parser, "while scanning a block scalar",
  2032. start_mark, "found a tab character where an indentation space is expected")
  2033. }
  2034. // Have we found a non-empty line?
  2035. if !is_break(parser.buffer, parser.buffer_pos) {
  2036. break
  2037. }
  2038. // Consume the line break.
  2039. if parser.unread < 2 && !yaml_parser_update_buffer(parser, 2) {
  2040. return false
  2041. }
  2042. // [Go] Should really be returning breaks instead.
  2043. *breaks = read_line(parser, *breaks)
  2044. *end_mark = parser.mark
  2045. }
  2046. // Determine the indentation level if needed.
  2047. if *indent == 0 {
  2048. *indent = max_indent
  2049. if *indent < parser.indent+1 {
  2050. *indent = parser.indent + 1
  2051. }
  2052. if *indent < 1 {
  2053. *indent = 1
  2054. }
  2055. }
  2056. return true
  2057. }
  2058. // Scan a quoted scalar.
  2059. func yaml_parser_scan_flow_scalar(parser *yaml_parser_t, token *yaml_token_t, single bool) bool {
  2060. // Eat the left quote.
  2061. start_mark := parser.mark
  2062. skip(parser)
  2063. // Consume the content of the quoted scalar.
  2064. var s, leading_break, trailing_breaks, whitespaces []byte
  2065. for {
  2066. // Check that there are no document indicators at the beginning of the line.
  2067. if parser.unread < 4 && !yaml_parser_update_buffer(parser, 4) {
  2068. return false
  2069. }
  2070. if parser.mark.column == 0 &&
  2071. ((parser.buffer[parser.buffer_pos+0] == '-' &&
  2072. parser.buffer[parser.buffer_pos+1] == '-' &&
  2073. parser.buffer[parser.buffer_pos+2] == '-') ||
  2074. (parser.buffer[parser.buffer_pos+0] == '.' &&
  2075. parser.buffer[parser.buffer_pos+1] == '.' &&
  2076. parser.buffer[parser.buffer_pos+2] == '.')) &&
  2077. is_blankz(parser.buffer, parser.buffer_pos+3) {
  2078. yaml_parser_set_scanner_error(parser, "while scanning a quoted scalar",
  2079. start_mark, "found unexpected document indicator")
  2080. return false
  2081. }
  2082. // Check for EOF.
  2083. if is_z(parser.buffer, parser.buffer_pos) {
  2084. yaml_parser_set_scanner_error(parser, "while scanning a quoted scalar",
  2085. start_mark, "found unexpected end of stream")
  2086. return false
  2087. }
  2088. // Consume non-blank characters.
  2089. leading_blanks := false
  2090. for !is_blankz(parser.buffer, parser.buffer_pos) {
  2091. if single && parser.buffer[parser.buffer_pos] == '\'' && parser.buffer[parser.buffer_pos+1] == '\'' {
  2092. // Is is an escaped single quote.
  2093. s = append(s, '\'')
  2094. skip(parser)
  2095. skip(parser)
  2096. } else if single && parser.buffer[parser.buffer_pos] == '\'' {
  2097. // It is a right single quote.
  2098. break
  2099. } else if !single && parser.buffer[parser.buffer_pos] == '"' {
  2100. // It is a right double quote.
  2101. break
  2102. } else if !single && parser.buffer[parser.buffer_pos] == '\\' && is_break(parser.buffer, parser.buffer_pos+1) {
  2103. // It is an escaped line break.
  2104. if parser.unread < 3 && !yaml_parser_update_buffer(parser, 3) {
  2105. return false
  2106. }
  2107. skip(parser)
  2108. skip_line(parser)
  2109. leading_blanks = true
  2110. break
  2111. } else if !single && parser.buffer[parser.buffer_pos] == '\\' {
  2112. // It is an escape sequence.
  2113. code_length := 0
  2114. // Check the escape character.
  2115. switch parser.buffer[parser.buffer_pos+1] {
  2116. case '0':
  2117. s = append(s, 0)
  2118. case 'a':
  2119. s = append(s, '\x07')
  2120. case 'b':
  2121. s = append(s, '\x08')
  2122. case 't', '\t':
  2123. s = append(s, '\x09')
  2124. case 'n':
  2125. s = append(s, '\x0A')
  2126. case 'v':
  2127. s = append(s, '\x0B')
  2128. case 'f':
  2129. s = append(s, '\x0C')
  2130. case 'r':
  2131. s = append(s, '\x0D')
  2132. case 'e':
  2133. s = append(s, '\x1B')
  2134. case ' ':
  2135. s = append(s, '\x20')
  2136. case '"':
  2137. s = append(s, '"')
  2138. case '\'':
  2139. s = append(s, '\'')
  2140. case '\\':
  2141. s = append(s, '\\')
  2142. case 'N': // NEL (#x85)
  2143. s = append(s, '\xC2')
  2144. s = append(s, '\x85')
  2145. case '_': // #xA0
  2146. s = append(s, '\xC2')
  2147. s = append(s, '\xA0')
  2148. case 'L': // LS (#x2028)
  2149. s = append(s, '\xE2')
  2150. s = append(s, '\x80')
  2151. s = append(s, '\xA8')
  2152. case 'P': // PS (#x2029)
  2153. s = append(s, '\xE2')
  2154. s = append(s, '\x80')
  2155. s = append(s, '\xA9')
  2156. case 'x':
  2157. code_length = 2
  2158. case 'u':
  2159. code_length = 4
  2160. case 'U':
  2161. code_length = 8
  2162. default:
  2163. yaml_parser_set_scanner_error(parser, "while parsing a quoted scalar",
  2164. start_mark, "found unknown escape character")
  2165. return false
  2166. }
  2167. skip(parser)
  2168. skip(parser)
  2169. // Consume an arbitrary escape code.
  2170. if code_length > 0 {
  2171. var value int
  2172. // Scan the character value.
  2173. if parser.unread < code_length && !yaml_parser_update_buffer(parser, code_length) {
  2174. return false
  2175. }
  2176. for k := 0; k < code_length; k++ {
  2177. if !is_hex(parser.buffer, parser.buffer_pos+k) {
  2178. yaml_parser_set_scanner_error(parser, "while parsing a quoted scalar",
  2179. start_mark, "did not find expected hexdecimal number")
  2180. return false
  2181. }
  2182. value = (value << 4) + as_hex(parser.buffer, parser.buffer_pos+k)
  2183. }
  2184. // Check the value and write the character.
  2185. if (value >= 0xD800 && value <= 0xDFFF) || value > 0x10FFFF {
  2186. yaml_parser_set_scanner_error(parser, "while parsing a quoted scalar",
  2187. start_mark, "found invalid Unicode character escape code")
  2188. return false
  2189. }
  2190. if value <= 0x7F {
  2191. s = append(s, byte(value))
  2192. } else if value <= 0x7FF {
  2193. s = append(s, byte(0xC0+(value>>6)))
  2194. s = append(s, byte(0x80+(value&0x3F)))
  2195. } else if value <= 0xFFFF {
  2196. s = append(s, byte(0xE0+(value>>12)))
  2197. s = append(s, byte(0x80+((value>>6)&0x3F)))
  2198. s = append(s, byte(0x80+(value&0x3F)))
  2199. } else {
  2200. s = append(s, byte(0xF0+(value>>18)))
  2201. s = append(s, byte(0x80+((value>>12)&0x3F)))
  2202. s = append(s, byte(0x80+((value>>6)&0x3F)))
  2203. s = append(s, byte(0x80+(value&0x3F)))
  2204. }
  2205. // Advance the pointer.
  2206. for k := 0; k < code_length; k++ {
  2207. skip(parser)
  2208. }
  2209. }
  2210. } else {
  2211. // It is a non-escaped non-blank character.
  2212. s = read(parser, s)
  2213. }
  2214. if parser.unread < 2 && !yaml_parser_update_buffer(parser, 2) {
  2215. return false
  2216. }
  2217. }
  2218. if parser.unread < 1 && !yaml_parser_update_buffer(parser, 1) {
  2219. return false
  2220. }
  2221. // Check if we are at the end of the scalar.
  2222. if single {
  2223. if parser.buffer[parser.buffer_pos] == '\'' {
  2224. break
  2225. }
  2226. } else {
  2227. if parser.buffer[parser.buffer_pos] == '"' {
  2228. break
  2229. }
  2230. }
  2231. // Consume blank characters.
  2232. for is_blank(parser.buffer, parser.buffer_pos) || is_break(parser.buffer, parser.buffer_pos) {
  2233. if is_blank(parser.buffer, parser.buffer_pos) {
  2234. // Consume a space or a tab character.
  2235. if !leading_blanks {
  2236. whitespaces = read(parser, whitespaces)
  2237. } else {
  2238. skip(parser)
  2239. }
  2240. } else {
  2241. if parser.unread < 2 && !yaml_parser_update_buffer(parser, 2) {
  2242. return false
  2243. }
  2244. // Check if it is a first line break.
  2245. if !leading_blanks {
  2246. whitespaces = whitespaces[:0]
  2247. leading_break = read_line(parser, leading_break)
  2248. leading_blanks = true
  2249. } else {
  2250. trailing_breaks = read_line(parser, trailing_breaks)
  2251. }
  2252. }
  2253. if parser.unread < 1 && !yaml_parser_update_buffer(parser, 1) {
  2254. return false
  2255. }
  2256. }
  2257. // Join the whitespaces or fold line breaks.
  2258. if leading_blanks {
  2259. // Do we need to fold line breaks?
  2260. if len(leading_break) > 0 && leading_break[0] == '\n' {
  2261. if len(trailing_breaks) == 0 {
  2262. s = append(s, ' ')
  2263. } else {
  2264. s = append(s, trailing_breaks...)
  2265. }
  2266. } else {
  2267. s = append(s, leading_break...)
  2268. s = append(s, trailing_breaks...)
  2269. }
  2270. trailing_breaks = trailing_breaks[:0]
  2271. leading_break = leading_break[:0]
  2272. } else {
  2273. s = append(s, whitespaces...)
  2274. whitespaces = whitespaces[:0]
  2275. }
  2276. }
  2277. // Eat the right quote.
  2278. skip(parser)
  2279. end_mark := parser.mark
  2280. // Create a token.
  2281. *token = yaml_token_t{
  2282. typ: yaml_SCALAR_TOKEN,
  2283. start_mark: start_mark,
  2284. end_mark: end_mark,
  2285. value: s,
  2286. style: yaml_SINGLE_QUOTED_SCALAR_STYLE,
  2287. }
  2288. if !single {
  2289. token.style = yaml_DOUBLE_QUOTED_SCALAR_STYLE
  2290. }
  2291. return true
  2292. }
  2293. // Scan a plain scalar.
  2294. func yaml_parser_scan_plain_scalar(parser *yaml_parser_t, token *yaml_token_t) bool {
  2295. var s, leading_break, trailing_breaks, whitespaces []byte
  2296. var leading_blanks bool
  2297. var indent = parser.indent + 1
  2298. start_mark := parser.mark
  2299. end_mark := parser.mark
  2300. // Consume the content of the plain scalar.
  2301. for {
  2302. // Check for a document indicator.
  2303. if parser.unread < 4 && !yaml_parser_update_buffer(parser, 4) {
  2304. return false
  2305. }
  2306. if parser.mark.column == 0 &&
  2307. ((parser.buffer[parser.buffer_pos+0] == '-' &&
  2308. parser.buffer[parser.buffer_pos+1] == '-' &&
  2309. parser.buffer[parser.buffer_pos+2] == '-') ||
  2310. (parser.buffer[parser.buffer_pos+0] == '.' &&
  2311. parser.buffer[parser.buffer_pos+1] == '.' &&
  2312. parser.buffer[parser.buffer_pos+2] == '.')) &&
  2313. is_blankz(parser.buffer, parser.buffer_pos+3) {
  2314. break
  2315. }
  2316. // Check for a comment.
  2317. if parser.buffer[parser.buffer_pos] == '#' {
  2318. break
  2319. }
  2320. // Consume non-blank characters.
  2321. for !is_blankz(parser.buffer, parser.buffer_pos) {
  2322. // Check for indicators that may end a plain scalar.
  2323. if (parser.buffer[parser.buffer_pos] == ':' && is_blankz(parser.buffer, parser.buffer_pos+1)) ||
  2324. (parser.flow_level > 0 &&
  2325. (parser.buffer[parser.buffer_pos] == ',' ||
  2326. parser.buffer[parser.buffer_pos] == '?' || parser.buffer[parser.buffer_pos] == '[' ||
  2327. parser.buffer[parser.buffer_pos] == ']' || parser.buffer[parser.buffer_pos] == '{' ||
  2328. parser.buffer[parser.buffer_pos] == '}')) {
  2329. break
  2330. }
  2331. // Check if we need to join whitespaces and breaks.
  2332. if leading_blanks || len(whitespaces) > 0 {
  2333. if leading_blanks {
  2334. // Do we need to fold line breaks?
  2335. if leading_break[0] == '\n' {
  2336. if len(trailing_breaks) == 0 {
  2337. s = append(s, ' ')
  2338. } else {
  2339. s = append(s, trailing_breaks...)
  2340. }
  2341. } else {
  2342. s = append(s, leading_break...)
  2343. s = append(s, trailing_breaks...)
  2344. }
  2345. trailing_breaks = trailing_breaks[:0]
  2346. leading_break = leading_break[:0]
  2347. leading_blanks = false
  2348. } else {
  2349. s = append(s, whitespaces...)
  2350. whitespaces = whitespaces[:0]
  2351. }
  2352. }
  2353. // Copy the character.
  2354. s = read(parser, s)
  2355. end_mark = parser.mark
  2356. if parser.unread < 2 && !yaml_parser_update_buffer(parser, 2) {
  2357. return false
  2358. }
  2359. }
  2360. // Is it the end?
  2361. if !(is_blank(parser.buffer, parser.buffer_pos) || is_break(parser.buffer, parser.buffer_pos)) {
  2362. break
  2363. }
  2364. // Consume blank characters.
  2365. if parser.unread < 1 && !yaml_parser_update_buffer(parser, 1) {
  2366. return false
  2367. }
  2368. for is_blank(parser.buffer, parser.buffer_pos) || is_break(parser.buffer, parser.buffer_pos) {
  2369. if is_blank(parser.buffer, parser.buffer_pos) {
  2370. // Check for tab characters that abuse indentation.
  2371. if leading_blanks && parser.mark.column < indent && is_tab(parser.buffer, parser.buffer_pos) {
  2372. yaml_parser_set_scanner_error(parser, "while scanning a plain scalar",
  2373. start_mark, "found a tab character that violates indentation")
  2374. return false
  2375. }
  2376. // Consume a space or a tab character.
  2377. if !leading_blanks {
  2378. whitespaces = read(parser, whitespaces)
  2379. } else {
  2380. skip(parser)
  2381. }
  2382. } else {
  2383. if parser.unread < 2 && !yaml_parser_update_buffer(parser, 2) {
  2384. return false
  2385. }
  2386. // Check if it is a first line break.
  2387. if !leading_blanks {
  2388. whitespaces = whitespaces[:0]
  2389. leading_break = read_line(parser, leading_break)
  2390. leading_blanks = true
  2391. } else {
  2392. trailing_breaks = read_line(parser, trailing_breaks)
  2393. }
  2394. }
  2395. if parser.unread < 1 && !yaml_parser_update_buffer(parser, 1) {
  2396. return false
  2397. }
  2398. }
  2399. // Check indentation level.
  2400. if parser.flow_level == 0 && parser.mark.column < indent {
  2401. break
  2402. }
  2403. }
  2404. // Create a token.
  2405. *token = yaml_token_t{
  2406. typ: yaml_SCALAR_TOKEN,
  2407. start_mark: start_mark,
  2408. end_mark: end_mark,
  2409. value: s,
  2410. style: yaml_PLAIN_SCALAR_STYLE,
  2411. }
  2412. // Note that we change the 'simple_key_allowed' flag.
  2413. if leading_blanks {
  2414. parser.simple_key_allowed = true
  2415. }
  2416. return true
  2417. }
  2418. func yaml_parser_scan_line_comment(parser *yaml_parser_t, after yaml_mark_t) bool {
  2419. if parser.mark.column == 0 {
  2420. return true
  2421. }
  2422. parser.comments = append(parser.comments, yaml_comment_t{after: after})
  2423. comment := &parser.comments[len(parser.comments)-1].line
  2424. for peek := 0; peek < 512; peek++ {
  2425. if parser.unread < peek+1 && !yaml_parser_update_buffer(parser, peek+1) {
  2426. break
  2427. }
  2428. if is_blank(parser.buffer, parser.buffer_pos+peek) {
  2429. continue
  2430. }
  2431. if parser.buffer[parser.buffer_pos+peek] == '#' {
  2432. if len(*comment) > 0 {
  2433. *comment = append(*comment, '\n')
  2434. }
  2435. for !is_breakz(parser.buffer, parser.buffer_pos+peek) {
  2436. *comment = append(*comment, parser.buffer[parser.buffer_pos+peek])
  2437. peek++
  2438. if parser.unread < peek+1 && !yaml_parser_update_buffer(parser, peek+1) {
  2439. return false
  2440. }
  2441. }
  2442. // Skip until after the consumed comment line.
  2443. until := parser.buffer_pos + peek
  2444. for parser.buffer_pos < until {
  2445. if is_break(parser.buffer, parser.buffer_pos) {
  2446. //break // Leave the break in the buffer so calling this function twice is safe.
  2447. if parser.unread < 2 && !yaml_parser_update_buffer(parser, 2) {
  2448. return false
  2449. }
  2450. skip_line(parser)
  2451. } else {
  2452. skip(parser)
  2453. }
  2454. }
  2455. }
  2456. break
  2457. }
  2458. return true
  2459. }
  2460. func yaml_parser_scan_head_comment(parser *yaml_parser_t, after yaml_mark_t) bool {
  2461. parser.comments = append(parser.comments, yaml_comment_t{after: after})
  2462. comment := &parser.comments[len(parser.comments)-1].head
  2463. breaks := false
  2464. for peek := 0; peek < 512; peek++ {
  2465. if parser.unread < peek+1 && !yaml_parser_update_buffer(parser, peek+1) {
  2466. break
  2467. }
  2468. if parser.buffer[parser.buffer_pos+peek] == 0 {
  2469. break
  2470. }
  2471. if is_blank(parser.buffer, parser.buffer_pos+peek) {
  2472. continue
  2473. }
  2474. if is_break(parser.buffer, parser.buffer_pos+peek) {
  2475. if !breaks {
  2476. *comment = append(*comment, '\n')
  2477. }
  2478. breaks = true
  2479. } else if parser.buffer[parser.buffer_pos+peek] == '#' {
  2480. if len(*comment) > 0 {
  2481. *comment = append(*comment, '\n')
  2482. }
  2483. breaks = false
  2484. for !is_breakz(parser.buffer, parser.buffer_pos+peek) {
  2485. *comment = append(*comment, parser.buffer[parser.buffer_pos+peek])
  2486. peek++
  2487. if parser.unread < peek+1 && !yaml_parser_update_buffer(parser, peek+1) {
  2488. return false
  2489. }
  2490. }
  2491. // Skip until after the consumed comment line.
  2492. until := parser.buffer_pos + peek
  2493. for parser.buffer_pos < until {
  2494. if is_break(parser.buffer, parser.buffer_pos) {
  2495. if parser.unread < 2 && !yaml_parser_update_buffer(parser, 2) {
  2496. return false
  2497. }
  2498. skip_line(parser)
  2499. } else {
  2500. skip(parser)
  2501. }
  2502. }
  2503. peek = 0
  2504. } else {
  2505. break
  2506. }
  2507. }
  2508. return true
  2509. }
  2510. func yaml_parser_scan_foot_comment(parser *yaml_parser_t, after yaml_mark_t) bool {
  2511. parser.comments = append(parser.comments, yaml_comment_t{after: after})
  2512. comment := &parser.comments[len(parser.comments)-1].foot
  2513. original := *comment
  2514. breaks := false
  2515. peek := 0
  2516. for ; peek < 32768; peek++ {
  2517. if parser.unread < peek+1 && !yaml_parser_update_buffer(parser, peek+1) {
  2518. break
  2519. }
  2520. c := parser.buffer[parser.buffer_pos+peek]
  2521. if c == 0 {
  2522. break
  2523. }
  2524. if is_blank(parser.buffer, parser.buffer_pos+peek) {
  2525. continue
  2526. }
  2527. if is_break(parser.buffer, parser.buffer_pos+peek) {
  2528. if breaks {
  2529. break
  2530. }
  2531. breaks = true
  2532. } else if c == '#' {
  2533. if len(*comment) > 0 {
  2534. *comment = append(*comment, '\n')
  2535. }
  2536. for !is_breakz(parser.buffer, parser.buffer_pos+peek) {
  2537. *comment = append(*comment, parser.buffer[parser.buffer_pos+peek])
  2538. peek++
  2539. if parser.unread < peek+1 && !yaml_parser_update_buffer(parser, peek+1) {
  2540. return false
  2541. }
  2542. }
  2543. breaks = true
  2544. } else if c == ']' || c == '}' {
  2545. break
  2546. } else {
  2547. // Abort and allow that next line to have the comment as its header.
  2548. *comment = original
  2549. return true
  2550. }
  2551. }
  2552. // Skip until after the consumed comment lines.
  2553. until := parser.buffer_pos + peek
  2554. for parser.buffer_pos < until {
  2555. if is_break(parser.buffer, parser.buffer_pos) {
  2556. if parser.unread < 2 && !yaml_parser_update_buffer(parser, 2) {
  2557. return false
  2558. }
  2559. skip_line(parser)
  2560. } else {
  2561. skip(parser)
  2562. }
  2563. }
  2564. return true
  2565. }