You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

PackParser.java 54KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323324325326327328329330331332333334335336337338339340341342343344345346347348349350351352353354355356357358359360361362363364365366367368369370371372373374375376377378379380381382383384385386387388389390391392393394395396397398399400401402403404405406407408409410411412413414415416417418419420421422423424425426427428429430431432433434435436437438439440441442443444445446447448449450451452453454455456457458459460461462463464465466467468469470471472473474475476477478479480481482483484485486487488489490491492493494495496497498499500501502503504505506507508509510511512513514515516517518519520521522523524525526527528529530531532533534535536537538539540541542543544545546547548549550551552553554555556557558559560561562563564565566567568569570571572573574575576577578579580581582583584585586587588589590591592593594595596597598599600601602603604605606607608609610611612613614615616617618619620621622623624625626627628629630631632633634635636637638639640641642643644645646647648649650651652653654655656657658659660661662663664665666667668669670671672673674675676677678679680681682683684685686687688689690691692693694695696697698699700701702703704705706707708709710711712713714715716717718719720721722723724725726727728729730731732733734735736737738739740741742743744745746747748749750751752753754755756757758759760761762763764765766767768769770771772773774775776777778779780781782783784785786787788789790791792793794795796797798799800801802803804805806807808809810811812813814815816817818819820821822823824825826827828829830831832833834835836837838839840841842843844845846847848849850851852853854855856857858859860861862863864865866867868869870871872873874875876877878879880881882883884885886887888889890891892893894895896897898899900901902903904905906907908909910911912913914915916917918919920921922923924925926927928929930931932933934935936937938939940941942943944945946947948949950951952953954955956957958959960961962963964965966967968969970971972973974975976977978979980981982983984985986987988989990991992993994995996997998999100010011002100310041005100610071008100910101011101210131014101510161017101810191020102110221023102410251026102710281029103010311032103310341035103610371038103910401041104210431044104510461047104810491050105110521053105410551056105710581059106010611062106310641065106610671068106910701071107210731074107510761077107810791080108110821083108410851086108710881089109010911092109310941095109610971098109911001101110211031104110511061107110811091110111111121113111411151116111711181119112011211122112311241125112611271128112911301131113211331134113511361137113811391140114111421143114411451146114711481149115011511152115311541155115611571158115911601161116211631164116511661167116811691170117111721173117411751176117711781179118011811182118311841185118611871188118911901191119211931194119511961197119811991200120112021203120412051206120712081209121012111212121312141215121612171218121912201221122212231224122512261227122812291230123112321233123412351236123712381239124012411242124312441245124612471248124912501251125212531254125512561257125812591260126112621263126412651266126712681269127012711272127312741275127612771278127912801281128212831284128512861287128812891290129112921293129412951296129712981299130013011302130313041305130613071308130913101311131213131314131513161317131813191320132113221323132413251326132713281329133013311332133313341335133613371338133913401341134213431344134513461347134813491350135113521353135413551356135713581359136013611362136313641365136613671368136913701371137213731374137513761377137813791380138113821383138413851386138713881389139013911392139313941395139613971398139914001401140214031404140514061407140814091410141114121413141414151416141714181419142014211422142314241425142614271428142914301431143214331434143514361437143814391440144114421443144414451446144714481449145014511452145314541455145614571458145914601461146214631464146514661467146814691470147114721473147414751476147714781479148014811482148314841485148614871488148914901491149214931494149514961497149814991500150115021503150415051506150715081509151015111512151315141515151615171518151915201521152215231524152515261527152815291530153115321533153415351536153715381539154015411542154315441545154615471548154915501551155215531554155515561557155815591560156115621563156415651566156715681569157015711572157315741575157615771578157915801581158215831584158515861587158815891590159115921593159415951596159715981599160016011602160316041605160616071608160916101611161216131614161516161617161816191620162116221623162416251626162716281629163016311632163316341635163616371638163916401641164216431644164516461647164816491650165116521653165416551656165716581659166016611662166316641665166616671668166916701671167216731674167516761677167816791680168116821683168416851686168716881689169016911692169316941695169616971698169917001701170217031704170517061707170817091710171117121713171417151716171717181719172017211722172317241725172617271728172917301731173217331734173517361737173817391740174117421743174417451746174717481749175017511752175317541755175617571758175917601761176217631764176517661767176817691770177117721773177417751776177717781779178017811782178317841785178617871788178917901791179217931794179517961797179817991800180118021803180418051806180718081809181018111812181318141815181618171818181918201821182218231824182518261827182818291830183118321833183418351836183718381839184018411842184318441845184618471848184918501851185218531854
  1. /*
  2. * Copyright (C) 2008-2011, Google Inc.
  3. * Copyright (C) 2007-2008, Robin Rosenberg <robin.rosenberg@dewire.com>
  4. * Copyright (C) 2008, Shawn O. Pearce <spearce@spearce.org>
  5. * and other copyright owners as documented in the project's IP log.
  6. *
  7. * This program and the accompanying materials are made available
  8. * under the terms of the Eclipse Distribution License v1.0 which
  9. * accompanies this distribution, is reproduced below, and is
  10. * available at http://www.eclipse.org/org/documents/edl-v10.php
  11. *
  12. * All rights reserved.
  13. *
  14. * Redistribution and use in source and binary forms, with or
  15. * without modification, are permitted provided that the following
  16. * conditions are met:
  17. *
  18. * - Redistributions of source code must retain the above copyright
  19. * notice, this list of conditions and the following disclaimer.
  20. *
  21. * - Redistributions in binary form must reproduce the above
  22. * copyright notice, this list of conditions and the following
  23. * disclaimer in the documentation and/or other materials provided
  24. * with the distribution.
  25. *
  26. * - Neither the name of the Eclipse Foundation, Inc. nor the
  27. * names of its contributors may be used to endorse or promote
  28. * products derived from this software without specific prior
  29. * written permission.
  30. *
  31. * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND
  32. * CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES,
  33. * INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
  34. * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
  35. * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR
  36. * CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
  37. * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
  38. * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
  39. * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
  40. * CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,
  41. * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
  42. * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF
  43. * ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
  44. */
  45. package org.eclipse.jgit.transport;
  46. import java.io.EOFException;
  47. import java.io.IOException;
  48. import java.io.InputStream;
  49. import java.security.MessageDigest;
  50. import java.text.MessageFormat;
  51. import java.util.ArrayList;
  52. import java.util.Arrays;
  53. import java.util.Comparator;
  54. import java.util.List;
  55. import java.util.concurrent.TimeUnit;
  56. import java.util.zip.DataFormatException;
  57. import java.util.zip.Inflater;
  58. import org.eclipse.jgit.errors.CorruptObjectException;
  59. import org.eclipse.jgit.errors.MissingObjectException;
  60. import org.eclipse.jgit.errors.TooLargeObjectInPackException;
  61. import org.eclipse.jgit.internal.JGitText;
  62. import org.eclipse.jgit.internal.storage.file.PackLock;
  63. import org.eclipse.jgit.internal.storage.pack.BinaryDelta;
  64. import org.eclipse.jgit.lib.AnyObjectId;
  65. import org.eclipse.jgit.lib.BatchingProgressMonitor;
  66. import org.eclipse.jgit.lib.BlobObjectChecker;
  67. import org.eclipse.jgit.lib.Constants;
  68. import org.eclipse.jgit.lib.InflaterCache;
  69. import org.eclipse.jgit.lib.MutableObjectId;
  70. import org.eclipse.jgit.lib.NullProgressMonitor;
  71. import org.eclipse.jgit.lib.ObjectChecker;
  72. import org.eclipse.jgit.lib.ObjectDatabase;
  73. import org.eclipse.jgit.lib.ObjectId;
  74. import org.eclipse.jgit.lib.ObjectIdOwnerMap;
  75. import org.eclipse.jgit.lib.ObjectIdSubclassMap;
  76. import org.eclipse.jgit.lib.ObjectLoader;
  77. import org.eclipse.jgit.lib.ObjectReader;
  78. import org.eclipse.jgit.lib.ObjectStream;
  79. import org.eclipse.jgit.lib.ProgressMonitor;
  80. import org.eclipse.jgit.util.BlockList;
  81. import org.eclipse.jgit.util.IO;
  82. import org.eclipse.jgit.util.LongMap;
  83. import org.eclipse.jgit.util.NB;
  84. import org.eclipse.jgit.util.sha1.SHA1;
  85. /**
  86. * Parses a pack stream and imports it for an
  87. * {@link org.eclipse.jgit.lib.ObjectInserter}.
  88. * <p>
  89. * Applications can acquire an instance of a parser from ObjectInserter's
  90. * {@link org.eclipse.jgit.lib.ObjectInserter#newPackParser(InputStream)}
  91. * method.
  92. * <p>
  93. * Implementations of {@link org.eclipse.jgit.lib.ObjectInserter} should
  94. * subclass this type and provide their own logic for the various {@code on*()}
  95. * event methods declared to be abstract.
  96. */
  97. public abstract class PackParser {
  98. /** Size of the internal stream buffer. */
  99. private static final int BUFFER_SIZE = 8192;
  100. /** Location data is being obtained from. */
  101. public static enum Source {
  102. /** Data is read from the incoming stream. */
  103. INPUT,
  104. /** Data is read back from the database's buffers. */
  105. DATABASE;
  106. }
  107. /** Object database used for loading existing objects. */
  108. private final ObjectDatabase objectDatabase;
  109. private InflaterStream inflater;
  110. private byte[] tempBuffer;
  111. private byte[] hdrBuf;
  112. private final SHA1 objectHasher = SHA1.newInstance();
  113. private final MutableObjectId tempObjectId;
  114. private InputStream in;
  115. byte[] buf;
  116. /** Position in the input stream of {@code buf[0]}. */
  117. private long bBase;
  118. private int bOffset;
  119. int bAvail;
  120. private ObjectChecker objCheck;
  121. private boolean allowThin;
  122. private boolean checkObjectCollisions;
  123. private boolean needBaseObjectIds;
  124. private boolean checkEofAfterPackFooter;
  125. private boolean expectDataAfterPackFooter;
  126. private long expectedObjectCount;
  127. private PackedObjectInfo[] entries;
  128. /**
  129. * Every object contained within the incoming pack.
  130. * <p>
  131. * This is a subset of {@link #entries}, as thin packs can add additional
  132. * objects to {@code entries} by copying already existing objects from the
  133. * repository onto the end of the thin pack to make it self-contained.
  134. */
  135. private ObjectIdSubclassMap<ObjectId> newObjectIds;
  136. private int deltaCount;
  137. private int entryCount;
  138. private ObjectIdOwnerMap<DeltaChain> baseById;
  139. /**
  140. * Objects referenced by their name from deltas, that aren't in this pack.
  141. * <p>
  142. * This is the set of objects that were copied onto the end of this pack to
  143. * make it complete. These objects were not transmitted by the remote peer,
  144. * but instead were assumed to already exist in the local repository.
  145. */
  146. private ObjectIdSubclassMap<ObjectId> baseObjectIds;
  147. private LongMap<UnresolvedDelta> baseByPos;
  148. /** Objects need to be double-checked for collision after indexing. */
  149. private BlockList<PackedObjectInfo> collisionCheckObjs;
  150. private MessageDigest packDigest;
  151. private ObjectReader readCurs;
  152. /** Message to protect the pack data from garbage collection. */
  153. private String lockMessage;
  154. /** Git object size limit */
  155. private long maxObjectSizeLimit;
  156. private final ReceivedPackStatistics.Builder stats =
  157. new ReceivedPackStatistics.Builder();
  158. /**
  159. * Initialize a pack parser.
  160. *
  161. * @param odb
  162. * database the parser will write its objects into.
  163. * @param src
  164. * the stream the parser will read.
  165. */
  166. protected PackParser(ObjectDatabase odb, InputStream src) {
  167. objectDatabase = odb.newCachedDatabase();
  168. in = src;
  169. inflater = new InflaterStream();
  170. readCurs = objectDatabase.newReader();
  171. buf = new byte[BUFFER_SIZE];
  172. tempBuffer = new byte[BUFFER_SIZE];
  173. hdrBuf = new byte[64];
  174. tempObjectId = new MutableObjectId();
  175. packDigest = Constants.newMessageDigest();
  176. checkObjectCollisions = true;
  177. }
  178. /**
  179. * Whether a thin pack (missing base objects) is permitted.
  180. *
  181. * @return {@code true} if a thin pack (missing base objects) is permitted.
  182. */
  183. public boolean isAllowThin() {
  184. return allowThin;
  185. }
  186. /**
  187. * Configure this index pack instance to allow a thin pack.
  188. * <p>
  189. * Thin packs are sometimes used during network transfers to allow a delta
  190. * to be sent without a base object. Such packs are not permitted on disk.
  191. *
  192. * @param allow
  193. * true to enable a thin pack.
  194. */
  195. public void setAllowThin(boolean allow) {
  196. allowThin = allow;
  197. }
  198. /**
  199. * Whether received objects are verified to prevent collisions.
  200. *
  201. * @return if true received objects are verified to prevent collisions.
  202. * @since 4.1
  203. */
  204. protected boolean isCheckObjectCollisions() {
  205. return checkObjectCollisions;
  206. }
  207. /**
  208. * Enable checking for collisions with existing objects.
  209. * <p>
  210. * By default PackParser looks for each received object in the repository.
  211. * If the object already exists, the existing object is compared
  212. * byte-for-byte with the newly received copy to ensure they are identical.
  213. * The receive is aborted with an exception if any byte differs. This check
  214. * is necessary to prevent an evil attacker from supplying a replacement
  215. * object into this repository in the event that a discovery enabling SHA-1
  216. * collisions is made.
  217. * <p>
  218. * This check may be very costly to perform, and some repositories may have
  219. * other ways to segregate newly received object data. The check is enabled
  220. * by default, but can be explicitly disabled if the implementation can
  221. * provide the same guarantee, or is willing to accept the risks associated
  222. * with bypassing the check.
  223. *
  224. * @param check
  225. * true to enable collision checking (strongly encouraged).
  226. * @since 4.1
  227. */
  228. protected void setCheckObjectCollisions(boolean check) {
  229. checkObjectCollisions = check;
  230. }
  231. /**
  232. * Configure this index pack instance to keep track of new objects.
  233. * <p>
  234. * By default an index pack doesn't save the new objects that were created
  235. * when it was instantiated. Setting this flag to {@code true} allows the
  236. * caller to use {@link #getNewObjectIds()} to retrieve that list.
  237. *
  238. * @param b
  239. * {@code true} to enable keeping track of new objects.
  240. */
  241. public void setNeedNewObjectIds(boolean b) {
  242. if (b)
  243. newObjectIds = new ObjectIdSubclassMap<>();
  244. else
  245. newObjectIds = null;
  246. }
  247. private boolean needNewObjectIds() {
  248. return newObjectIds != null;
  249. }
  250. /**
  251. * Configure this index pack instance to keep track of the objects assumed
  252. * for delta bases.
  253. * <p>
  254. * By default an index pack doesn't save the objects that were used as delta
  255. * bases. Setting this flag to {@code true} will allow the caller to use
  256. * {@link #getBaseObjectIds()} to retrieve that list.
  257. *
  258. * @param b
  259. * {@code true} to enable keeping track of delta bases.
  260. */
  261. public void setNeedBaseObjectIds(boolean b) {
  262. this.needBaseObjectIds = b;
  263. }
  264. /**
  265. * Whether the EOF should be read from the input after the footer.
  266. *
  267. * @return true if the EOF should be read from the input after the footer.
  268. */
  269. public boolean isCheckEofAfterPackFooter() {
  270. return checkEofAfterPackFooter;
  271. }
  272. /**
  273. * Ensure EOF is read from the input stream after the footer.
  274. *
  275. * @param b
  276. * true if the EOF should be read; false if it is not checked.
  277. */
  278. public void setCheckEofAfterPackFooter(boolean b) {
  279. checkEofAfterPackFooter = b;
  280. }
  281. /**
  282. * Whether there is data expected after the pack footer.
  283. *
  284. * @return true if there is data expected after the pack footer.
  285. */
  286. public boolean isExpectDataAfterPackFooter() {
  287. return expectDataAfterPackFooter;
  288. }
  289. /**
  290. * Set if there is additional data in InputStream after pack.
  291. *
  292. * @param e
  293. * true if there is additional data in InputStream after pack.
  294. * This requires the InputStream to support the mark and reset
  295. * functions.
  296. */
  297. public void setExpectDataAfterPackFooter(boolean e) {
  298. expectDataAfterPackFooter = e;
  299. }
  300. /**
  301. * Get the new objects that were sent by the user
  302. *
  303. * @return the new objects that were sent by the user
  304. */
  305. public ObjectIdSubclassMap<ObjectId> getNewObjectIds() {
  306. if (newObjectIds != null)
  307. return newObjectIds;
  308. return new ObjectIdSubclassMap<>();
  309. }
  310. /**
  311. * Get set of objects the incoming pack assumed for delta purposes
  312. *
  313. * @return set of objects the incoming pack assumed for delta purposes
  314. */
  315. public ObjectIdSubclassMap<ObjectId> getBaseObjectIds() {
  316. if (baseObjectIds != null)
  317. return baseObjectIds;
  318. return new ObjectIdSubclassMap<>();
  319. }
  320. /**
  321. * Configure the checker used to validate received objects.
  322. * <p>
  323. * Usually object checking isn't necessary, as Git implementations only
  324. * create valid objects in pack files. However, additional checking may be
  325. * useful if processing data from an untrusted source.
  326. *
  327. * @param oc
  328. * the checker instance; null to disable object checking.
  329. */
  330. public void setObjectChecker(ObjectChecker oc) {
  331. objCheck = oc;
  332. }
  333. /**
  334. * Configure the checker used to validate received objects.
  335. * <p>
  336. * Usually object checking isn't necessary, as Git implementations only
  337. * create valid objects in pack files. However, additional checking may be
  338. * useful if processing data from an untrusted source.
  339. * <p>
  340. * This is shorthand for:
  341. *
  342. * <pre>
  343. * setObjectChecker(on ? new ObjectChecker() : null);
  344. * </pre>
  345. *
  346. * @param on
  347. * true to enable the default checker; false to disable it.
  348. */
  349. public void setObjectChecking(boolean on) {
  350. setObjectChecker(on ? new ObjectChecker() : null);
  351. }
  352. /**
  353. * Get the message to record with the pack lock.
  354. *
  355. * @return the message to record with the pack lock.
  356. */
  357. public String getLockMessage() {
  358. return lockMessage;
  359. }
  360. /**
  361. * Set the lock message for the incoming pack data.
  362. *
  363. * @param msg
  364. * if not null, the message to associate with the incoming data
  365. * while it is locked to prevent garbage collection.
  366. */
  367. public void setLockMessage(String msg) {
  368. lockMessage = msg;
  369. }
  370. /**
  371. * Set the maximum allowed Git object size.
  372. * <p>
  373. * If an object is larger than the given size the pack-parsing will throw an
  374. * exception aborting the parsing.
  375. *
  376. * @param limit
  377. * the Git object size limit. If zero then there is not limit.
  378. */
  379. public void setMaxObjectSizeLimit(long limit) {
  380. maxObjectSizeLimit = limit;
  381. }
  382. /**
  383. * Get the number of objects in the stream.
  384. * <p>
  385. * The object count is only available after {@link #parse(ProgressMonitor)}
  386. * has returned. The count may have been increased if the stream was a thin
  387. * pack, and missing bases objects were appending onto it by the subclass.
  388. *
  389. * @return number of objects parsed out of the stream.
  390. */
  391. public int getObjectCount() {
  392. return entryCount;
  393. }
  394. /**
  395. * Get the information about the requested object.
  396. * <p>
  397. * The object information is only available after
  398. * {@link #parse(ProgressMonitor)} has returned.
  399. *
  400. * @param nth
  401. * index of the object in the stream. Must be between 0 and
  402. * {@link #getObjectCount()}-1.
  403. * @return the object information.
  404. */
  405. public PackedObjectInfo getObject(int nth) {
  406. return entries[nth];
  407. }
  408. /**
  409. * Get all of the objects, sorted by their name.
  410. * <p>
  411. * The object information is only available after
  412. * {@link #parse(ProgressMonitor)} has returned.
  413. * <p>
  414. * To maintain lower memory usage and good runtime performance, this method
  415. * sorts the objects in-place and therefore impacts the ordering presented
  416. * by {@link #getObject(int)}.
  417. *
  418. * @param cmp
  419. * comparison function, if null objects are stored by ObjectId.
  420. * @return sorted list of objects in this pack stream.
  421. */
  422. public List<PackedObjectInfo> getSortedObjectList(
  423. Comparator<PackedObjectInfo> cmp) {
  424. Arrays.sort(entries, 0, entryCount, cmp);
  425. List<PackedObjectInfo> list = Arrays.asList(entries);
  426. if (entryCount < entries.length)
  427. list = list.subList(0, entryCount);
  428. return list;
  429. }
  430. /**
  431. * Get the size of the newly created pack.
  432. * <p>
  433. * This will also include the pack index size if an index was created. This
  434. * method should only be called after pack parsing is finished.
  435. *
  436. * @return the pack size (including the index size) or -1 if the size cannot
  437. * be determined
  438. * @since 3.3
  439. */
  440. public long getPackSize() {
  441. return -1;
  442. }
  443. /**
  444. * Returns the statistics of the parsed pack.
  445. * <p>
  446. * This should only be called after pack parsing is finished.
  447. *
  448. * @return {@link org.eclipse.jgit.transport.ReceivedPackStatistics}
  449. * @since 4.6
  450. */
  451. public ReceivedPackStatistics getReceivedPackStatistics() {
  452. return stats.build();
  453. }
  454. /**
  455. * Parse the pack stream.
  456. *
  457. * @param progress
  458. * callback to provide progress feedback during parsing. If null,
  459. * {@link org.eclipse.jgit.lib.NullProgressMonitor} will be used.
  460. * @return the pack lock, if one was requested by setting
  461. * {@link #setLockMessage(String)}.
  462. * @throws java.io.IOException
  463. * the stream is malformed, or contains corrupt objects.
  464. * @since 3.0
  465. */
  466. public final PackLock parse(ProgressMonitor progress) throws IOException {
  467. return parse(progress, progress);
  468. }
  469. /**
  470. * Parse the pack stream.
  471. *
  472. * @param receiving
  473. * receives progress feedback during the initial receiving
  474. * objects phase. If null,
  475. * {@link org.eclipse.jgit.lib.NullProgressMonitor} will be used.
  476. * @param resolving
  477. * receives progress feedback during the resolving objects phase.
  478. * @return the pack lock, if one was requested by setting
  479. * {@link #setLockMessage(String)}.
  480. * @throws java.io.IOException
  481. * the stream is malformed, or contains corrupt objects.
  482. * @since 3.0
  483. */
  484. public PackLock parse(ProgressMonitor receiving, ProgressMonitor resolving)
  485. throws IOException {
  486. if (receiving == null)
  487. receiving = NullProgressMonitor.INSTANCE;
  488. if (resolving == null)
  489. resolving = NullProgressMonitor.INSTANCE;
  490. if (receiving == resolving)
  491. receiving.start(2 /* tasks */);
  492. try {
  493. readPackHeader();
  494. entries = new PackedObjectInfo[(int) expectedObjectCount];
  495. baseById = new ObjectIdOwnerMap<>();
  496. baseByPos = new LongMap<>();
  497. collisionCheckObjs = new BlockList<>();
  498. receiving.beginTask(JGitText.get().receivingObjects,
  499. (int) expectedObjectCount);
  500. try {
  501. for (int done = 0; done < expectedObjectCount; done++) {
  502. indexOneObject();
  503. receiving.update(1);
  504. if (receiving.isCancelled())
  505. throw new IOException(JGitText.get().downloadCancelled);
  506. }
  507. readPackFooter();
  508. endInput();
  509. } finally {
  510. receiving.endTask();
  511. }
  512. if (!collisionCheckObjs.isEmpty()) {
  513. checkObjectCollision();
  514. }
  515. if (deltaCount > 0) {
  516. processDeltas(resolving);
  517. }
  518. packDigest = null;
  519. baseById = null;
  520. baseByPos = null;
  521. } finally {
  522. try {
  523. if (readCurs != null)
  524. readCurs.close();
  525. } finally {
  526. readCurs = null;
  527. }
  528. try {
  529. inflater.release();
  530. } finally {
  531. inflater = null;
  532. }
  533. }
  534. return null; // By default there is no locking.
  535. }
  536. private void processDeltas(ProgressMonitor resolving) throws IOException {
  537. if (resolving instanceof BatchingProgressMonitor) {
  538. ((BatchingProgressMonitor) resolving).setDelayStart(1000,
  539. TimeUnit.MILLISECONDS);
  540. }
  541. resolving.beginTask(JGitText.get().resolvingDeltas, deltaCount);
  542. resolveDeltas(resolving);
  543. if (entryCount < expectedObjectCount) {
  544. if (!isAllowThin()) {
  545. throw new IOException(MessageFormat.format(
  546. JGitText.get().packHasUnresolvedDeltas,
  547. Long.valueOf(expectedObjectCount - entryCount)));
  548. }
  549. resolveDeltasWithExternalBases(resolving);
  550. if (entryCount < expectedObjectCount) {
  551. throw new IOException(MessageFormat.format(
  552. JGitText.get().packHasUnresolvedDeltas,
  553. Long.valueOf(expectedObjectCount - entryCount)));
  554. }
  555. }
  556. resolving.endTask();
  557. }
  558. private void resolveDeltas(ProgressMonitor progress)
  559. throws IOException {
  560. final int last = entryCount;
  561. for (int i = 0; i < last; i++) {
  562. resolveDeltas(entries[i], progress);
  563. if (progress.isCancelled())
  564. throw new IOException(
  565. JGitText.get().downloadCancelledDuringIndexing);
  566. }
  567. }
  568. private void resolveDeltas(final PackedObjectInfo oe,
  569. ProgressMonitor progress) throws IOException {
  570. UnresolvedDelta children = firstChildOf(oe);
  571. if (children == null)
  572. return;
  573. DeltaVisit visit = new DeltaVisit();
  574. visit.nextChild = children;
  575. ObjectTypeAndSize info = openDatabase(oe, new ObjectTypeAndSize());
  576. switch (info.type) {
  577. case Constants.OBJ_COMMIT:
  578. case Constants.OBJ_TREE:
  579. case Constants.OBJ_BLOB:
  580. case Constants.OBJ_TAG:
  581. visit.data = inflateAndReturn(Source.DATABASE, info.size);
  582. visit.id = oe;
  583. break;
  584. default:
  585. throw new IOException(MessageFormat.format(
  586. JGitText.get().unknownObjectType,
  587. Integer.valueOf(info.type)));
  588. }
  589. if (!checkCRC(oe.getCRC())) {
  590. throw new IOException(MessageFormat.format(
  591. JGitText.get().corruptionDetectedReReadingAt,
  592. Long.valueOf(oe.getOffset())));
  593. }
  594. resolveDeltas(visit.next(), info.type, info, progress);
  595. }
  596. private void resolveDeltas(DeltaVisit visit, final int type,
  597. ObjectTypeAndSize info, ProgressMonitor progress)
  598. throws IOException {
  599. stats.addDeltaObject(type);
  600. do {
  601. progress.update(1);
  602. info = openDatabase(visit.delta, info);
  603. switch (info.type) {
  604. case Constants.OBJ_OFS_DELTA:
  605. case Constants.OBJ_REF_DELTA:
  606. break;
  607. default:
  608. throw new IOException(MessageFormat.format(
  609. JGitText.get().unknownObjectType,
  610. Integer.valueOf(info.type)));
  611. }
  612. byte[] delta = inflateAndReturn(Source.DATABASE, info.size);
  613. checkIfTooLarge(type, BinaryDelta.getResultSize(delta));
  614. visit.data = BinaryDelta.apply(visit.parent.data, delta);
  615. delta = null;
  616. if (!checkCRC(visit.delta.crc))
  617. throw new IOException(MessageFormat.format(
  618. JGitText.get().corruptionDetectedReReadingAt,
  619. Long.valueOf(visit.delta.position)));
  620. SHA1 objectDigest = objectHasher.reset();
  621. objectDigest.update(Constants.encodedTypeString(type));
  622. objectDigest.update((byte) ' ');
  623. objectDigest.update(Constants.encodeASCII(visit.data.length));
  624. objectDigest.update((byte) 0);
  625. objectDigest.update(visit.data);
  626. objectDigest.digest(tempObjectId);
  627. verifySafeObject(tempObjectId, type, visit.data);
  628. if (isCheckObjectCollisions() && readCurs.has(tempObjectId)) {
  629. checkObjectCollision(tempObjectId, type, visit.data);
  630. }
  631. PackedObjectInfo oe;
  632. oe = newInfo(tempObjectId, visit.delta, visit.parent.id);
  633. oe.setOffset(visit.delta.position);
  634. oe.setType(type);
  635. onInflatedObjectData(oe, type, visit.data);
  636. addObjectAndTrack(oe);
  637. visit.id = oe;
  638. visit.nextChild = firstChildOf(oe);
  639. visit = visit.next();
  640. } while (visit != null);
  641. }
  642. private final void checkIfTooLarge(int typeCode, long size)
  643. throws IOException {
  644. if (0 < maxObjectSizeLimit && maxObjectSizeLimit < size) {
  645. switch (typeCode) {
  646. case Constants.OBJ_COMMIT:
  647. case Constants.OBJ_TREE:
  648. case Constants.OBJ_BLOB:
  649. case Constants.OBJ_TAG:
  650. throw new TooLargeObjectInPackException(size, maxObjectSizeLimit);
  651. case Constants.OBJ_OFS_DELTA:
  652. case Constants.OBJ_REF_DELTA:
  653. throw new TooLargeObjectInPackException(size, maxObjectSizeLimit);
  654. default:
  655. throw new IOException(MessageFormat.format(
  656. JGitText.get().unknownObjectType,
  657. Integer.valueOf(typeCode)));
  658. }
  659. }
  660. if (size > Integer.MAX_VALUE - 8) {
  661. throw new TooLargeObjectInPackException(size, Integer.MAX_VALUE - 8);
  662. }
  663. }
  664. /**
  665. * Read the header of the current object.
  666. * <p>
  667. * After the header has been parsed, this method automatically invokes
  668. * {@link #onObjectHeader(Source, byte[], int, int)} to allow the
  669. * implementation to update its internal checksums for the bytes read.
  670. * <p>
  671. * When this method returns the database will be positioned on the first
  672. * byte of the deflated data stream.
  673. *
  674. * @param info
  675. * the info object to populate.
  676. * @return {@code info}, after populating.
  677. * @throws java.io.IOException
  678. * the size cannot be read.
  679. */
  680. protected ObjectTypeAndSize readObjectHeader(ObjectTypeAndSize info)
  681. throws IOException {
  682. int hdrPtr = 0;
  683. int c = readFrom(Source.DATABASE);
  684. hdrBuf[hdrPtr++] = (byte) c;
  685. info.type = (c >> 4) & 7;
  686. long sz = c & 15;
  687. int shift = 4;
  688. while ((c & 0x80) != 0) {
  689. c = readFrom(Source.DATABASE);
  690. hdrBuf[hdrPtr++] = (byte) c;
  691. sz += ((long) (c & 0x7f)) << shift;
  692. shift += 7;
  693. }
  694. info.size = sz;
  695. switch (info.type) {
  696. case Constants.OBJ_COMMIT:
  697. case Constants.OBJ_TREE:
  698. case Constants.OBJ_BLOB:
  699. case Constants.OBJ_TAG:
  700. onObjectHeader(Source.DATABASE, hdrBuf, 0, hdrPtr);
  701. break;
  702. case Constants.OBJ_OFS_DELTA:
  703. c = readFrom(Source.DATABASE);
  704. hdrBuf[hdrPtr++] = (byte) c;
  705. while ((c & 128) != 0) {
  706. c = readFrom(Source.DATABASE);
  707. hdrBuf[hdrPtr++] = (byte) c;
  708. }
  709. onObjectHeader(Source.DATABASE, hdrBuf, 0, hdrPtr);
  710. break;
  711. case Constants.OBJ_REF_DELTA:
  712. System.arraycopy(buf, fill(Source.DATABASE, 20), hdrBuf, hdrPtr, 20);
  713. hdrPtr += 20;
  714. use(20);
  715. onObjectHeader(Source.DATABASE, hdrBuf, 0, hdrPtr);
  716. break;
  717. default:
  718. throw new IOException(MessageFormat.format(
  719. JGitText.get().unknownObjectType,
  720. Integer.valueOf(info.type)));
  721. }
  722. return info;
  723. }
  724. private UnresolvedDelta removeBaseById(AnyObjectId id) {
  725. final DeltaChain d = baseById.get(id);
  726. return d != null ? d.remove() : null;
  727. }
  728. private static UnresolvedDelta reverse(UnresolvedDelta c) {
  729. UnresolvedDelta tail = null;
  730. while (c != null) {
  731. final UnresolvedDelta n = c.next;
  732. c.next = tail;
  733. tail = c;
  734. c = n;
  735. }
  736. return tail;
  737. }
  738. private UnresolvedDelta firstChildOf(PackedObjectInfo oe) {
  739. UnresolvedDelta a = reverse(removeBaseById(oe));
  740. UnresolvedDelta b = reverse(baseByPos.remove(oe.getOffset()));
  741. if (a == null)
  742. return b;
  743. if (b == null)
  744. return a;
  745. UnresolvedDelta first = null;
  746. UnresolvedDelta last = null;
  747. while (a != null || b != null) {
  748. UnresolvedDelta curr;
  749. if (b == null || (a != null && a.position < b.position)) {
  750. curr = a;
  751. a = a.next;
  752. } else {
  753. curr = b;
  754. b = b.next;
  755. }
  756. if (last != null)
  757. last.next = curr;
  758. else
  759. first = curr;
  760. last = curr;
  761. curr.next = null;
  762. }
  763. return first;
  764. }
  765. private void resolveDeltasWithExternalBases(ProgressMonitor progress)
  766. throws IOException {
  767. growEntries(baseById.size());
  768. if (needBaseObjectIds)
  769. baseObjectIds = new ObjectIdSubclassMap<>();
  770. final List<DeltaChain> missing = new ArrayList<>(64);
  771. for (DeltaChain baseId : baseById) {
  772. if (baseId.head == null)
  773. continue;
  774. if (needBaseObjectIds)
  775. baseObjectIds.add(baseId);
  776. final ObjectLoader ldr;
  777. try {
  778. ldr = readCurs.open(baseId);
  779. } catch (MissingObjectException notFound) {
  780. missing.add(baseId);
  781. continue;
  782. }
  783. final DeltaVisit visit = new DeltaVisit();
  784. visit.data = ldr.getCachedBytes(Integer.MAX_VALUE);
  785. visit.id = baseId;
  786. final int typeCode = ldr.getType();
  787. final PackedObjectInfo oe = newInfo(baseId, null, null);
  788. oe.setType(typeCode);
  789. if (onAppendBase(typeCode, visit.data, oe))
  790. entries[entryCount++] = oe;
  791. visit.nextChild = firstChildOf(oe);
  792. resolveDeltas(visit.next(), typeCode,
  793. new ObjectTypeAndSize(), progress);
  794. if (progress.isCancelled())
  795. throw new IOException(
  796. JGitText.get().downloadCancelledDuringIndexing);
  797. }
  798. for (DeltaChain base : missing) {
  799. if (base.head != null)
  800. throw new MissingObjectException(base, "delta base"); //$NON-NLS-1$
  801. }
  802. onEndThinPack();
  803. }
  804. private void growEntries(int extraObjects) {
  805. final PackedObjectInfo[] ne;
  806. ne = new PackedObjectInfo[(int) expectedObjectCount + extraObjects];
  807. System.arraycopy(entries, 0, ne, 0, entryCount);
  808. entries = ne;
  809. }
  810. private void readPackHeader() throws IOException {
  811. if (expectDataAfterPackFooter) {
  812. if (!in.markSupported())
  813. throw new IOException(
  814. JGitText.get().inputStreamMustSupportMark);
  815. in.mark(buf.length);
  816. }
  817. final int hdrln = Constants.PACK_SIGNATURE.length + 4 + 4;
  818. final int p = fill(Source.INPUT, hdrln);
  819. for (int k = 0; k < Constants.PACK_SIGNATURE.length; k++)
  820. if (buf[p + k] != Constants.PACK_SIGNATURE[k])
  821. throw new IOException(JGitText.get().notAPACKFile);
  822. final long vers = NB.decodeUInt32(buf, p + 4);
  823. if (vers != 2 && vers != 3)
  824. throw new IOException(MessageFormat.format(
  825. JGitText.get().unsupportedPackVersion, Long.valueOf(vers)));
  826. final long objectCount = NB.decodeUInt32(buf, p + 8);
  827. use(hdrln);
  828. setExpectedObjectCount(objectCount);
  829. onPackHeader(objectCount);
  830. }
  831. private void readPackFooter() throws IOException {
  832. sync();
  833. final byte[] actHash = packDigest.digest();
  834. final int c = fill(Source.INPUT, 20);
  835. final byte[] srcHash = new byte[20];
  836. System.arraycopy(buf, c, srcHash, 0, 20);
  837. use(20);
  838. if (bAvail != 0 && !expectDataAfterPackFooter)
  839. throw new CorruptObjectException(MessageFormat.format(
  840. JGitText.get().expectedEOFReceived,
  841. "\\x" + Integer.toHexString(buf[bOffset] & 0xff))); //$NON-NLS-1$
  842. if (isCheckEofAfterPackFooter()) {
  843. int eof = in.read();
  844. if (0 <= eof)
  845. throw new CorruptObjectException(MessageFormat.format(
  846. JGitText.get().expectedEOFReceived,
  847. "\\x" + Integer.toHexString(eof))); //$NON-NLS-1$
  848. } else if (bAvail > 0 && expectDataAfterPackFooter) {
  849. in.reset();
  850. IO.skipFully(in, bOffset);
  851. }
  852. if (!Arrays.equals(actHash, srcHash))
  853. throw new CorruptObjectException(
  854. JGitText.get().corruptObjectPackfileChecksumIncorrect);
  855. onPackFooter(srcHash);
  856. }
  857. // Cleanup all resources associated with our input parsing.
  858. private void endInput() {
  859. stats.setNumBytesRead(streamPosition());
  860. in = null;
  861. }
  862. // Read one entire object or delta from the input.
  863. private void indexOneObject() throws IOException {
  864. final long streamPosition = streamPosition();
  865. int hdrPtr = 0;
  866. int c = readFrom(Source.INPUT);
  867. hdrBuf[hdrPtr++] = (byte) c;
  868. final int typeCode = (c >> 4) & 7;
  869. long sz = c & 15;
  870. int shift = 4;
  871. while ((c & 0x80) != 0) {
  872. c = readFrom(Source.INPUT);
  873. hdrBuf[hdrPtr++] = (byte) c;
  874. sz += ((long) (c & 0x7f)) << shift;
  875. shift += 7;
  876. }
  877. checkIfTooLarge(typeCode, sz);
  878. switch (typeCode) {
  879. case Constants.OBJ_COMMIT:
  880. case Constants.OBJ_TREE:
  881. case Constants.OBJ_BLOB:
  882. case Constants.OBJ_TAG:
  883. stats.addWholeObject(typeCode);
  884. onBeginWholeObject(streamPosition, typeCode, sz);
  885. onObjectHeader(Source.INPUT, hdrBuf, 0, hdrPtr);
  886. whole(streamPosition, typeCode, sz);
  887. break;
  888. case Constants.OBJ_OFS_DELTA: {
  889. stats.addOffsetDelta();
  890. c = readFrom(Source.INPUT);
  891. hdrBuf[hdrPtr++] = (byte) c;
  892. long ofs = c & 127;
  893. while ((c & 128) != 0) {
  894. ofs += 1;
  895. c = readFrom(Source.INPUT);
  896. hdrBuf[hdrPtr++] = (byte) c;
  897. ofs <<= 7;
  898. ofs += (c & 127);
  899. }
  900. final long base = streamPosition - ofs;
  901. onBeginOfsDelta(streamPosition, base, sz);
  902. onObjectHeader(Source.INPUT, hdrBuf, 0, hdrPtr);
  903. inflateAndSkip(Source.INPUT, sz);
  904. UnresolvedDelta n = onEndDelta();
  905. n.position = streamPosition;
  906. n.next = baseByPos.put(base, n);
  907. deltaCount++;
  908. break;
  909. }
  910. case Constants.OBJ_REF_DELTA: {
  911. stats.addRefDelta();
  912. c = fill(Source.INPUT, 20);
  913. final ObjectId base = ObjectId.fromRaw(buf, c);
  914. System.arraycopy(buf, c, hdrBuf, hdrPtr, 20);
  915. hdrPtr += 20;
  916. use(20);
  917. DeltaChain r = baseById.get(base);
  918. if (r == null) {
  919. r = new DeltaChain(base);
  920. baseById.add(r);
  921. }
  922. onBeginRefDelta(streamPosition, base, sz);
  923. onObjectHeader(Source.INPUT, hdrBuf, 0, hdrPtr);
  924. inflateAndSkip(Source.INPUT, sz);
  925. UnresolvedDelta n = onEndDelta();
  926. n.position = streamPosition;
  927. r.add(n);
  928. deltaCount++;
  929. break;
  930. }
  931. default:
  932. throw new IOException(
  933. MessageFormat.format(JGitText.get().unknownObjectType,
  934. Integer.valueOf(typeCode)));
  935. }
  936. }
  937. private void whole(long pos, int type, long sz)
  938. throws IOException {
  939. SHA1 objectDigest = objectHasher.reset();
  940. objectDigest.update(Constants.encodedTypeString(type));
  941. objectDigest.update((byte) ' ');
  942. objectDigest.update(Constants.encodeASCII(sz));
  943. objectDigest.update((byte) 0);
  944. final byte[] data;
  945. if (type == Constants.OBJ_BLOB) {
  946. byte[] readBuffer = buffer();
  947. BlobObjectChecker checker = null;
  948. if (objCheck != null) {
  949. checker = objCheck.newBlobObjectChecker();
  950. }
  951. if (checker == null) {
  952. checker = BlobObjectChecker.NULL_CHECKER;
  953. }
  954. long cnt = 0;
  955. try (InputStream inf = inflate(Source.INPUT, sz)) {
  956. while (cnt < sz) {
  957. int r = inf.read(readBuffer);
  958. if (r <= 0)
  959. break;
  960. objectDigest.update(readBuffer, 0, r);
  961. checker.update(readBuffer, 0, r);
  962. cnt += r;
  963. }
  964. }
  965. objectDigest.digest(tempObjectId);
  966. checker.endBlob(tempObjectId);
  967. data = null;
  968. } else {
  969. data = inflateAndReturn(Source.INPUT, sz);
  970. objectDigest.update(data);
  971. objectDigest.digest(tempObjectId);
  972. verifySafeObject(tempObjectId, type, data);
  973. }
  974. PackedObjectInfo obj = newInfo(tempObjectId, null, null);
  975. obj.setOffset(pos);
  976. obj.setType(type);
  977. onEndWholeObject(obj);
  978. if (data != null)
  979. onInflatedObjectData(obj, type, data);
  980. addObjectAndTrack(obj);
  981. if (isCheckObjectCollisions()) {
  982. collisionCheckObjs.add(obj);
  983. }
  984. }
  985. /**
  986. * Verify the integrity of the object.
  987. *
  988. * @param id
  989. * identity of the object to be checked.
  990. * @param type
  991. * the type of the object.
  992. * @param data
  993. * raw content of the object.
  994. * @throws org.eclipse.jgit.errors.CorruptObjectException
  995. * @since 4.9
  996. */
  997. protected void verifySafeObject(final AnyObjectId id, final int type,
  998. final byte[] data) throws CorruptObjectException {
  999. if (objCheck != null) {
  1000. try {
  1001. objCheck.check(id, type, data);
  1002. } catch (CorruptObjectException e) {
  1003. if (e.getErrorType() != null) {
  1004. throw e;
  1005. }
  1006. throw new CorruptObjectException(
  1007. MessageFormat.format(JGitText.get().invalidObject,
  1008. Constants.typeString(type), id.name(),
  1009. e.getMessage()),
  1010. e);
  1011. }
  1012. }
  1013. }
  1014. private void checkObjectCollision() throws IOException {
  1015. for (PackedObjectInfo obj : collisionCheckObjs) {
  1016. if (!readCurs.has(obj)) {
  1017. continue;
  1018. }
  1019. checkObjectCollision(obj);
  1020. }
  1021. }
  1022. private void checkObjectCollision(PackedObjectInfo obj)
  1023. throws IOException {
  1024. ObjectTypeAndSize info = openDatabase(obj, new ObjectTypeAndSize());
  1025. final byte[] readBuffer = buffer();
  1026. final byte[] curBuffer = new byte[readBuffer.length];
  1027. long sz = info.size;
  1028. try (ObjectStream cur = readCurs.open(obj, info.type).openStream()) {
  1029. if (cur.getSize() != sz) {
  1030. throw new IOException(MessageFormat.format(
  1031. JGitText.get().collisionOn, obj.name()));
  1032. }
  1033. try (InputStream pck = inflate(Source.DATABASE, sz)) {
  1034. while (0 < sz) {
  1035. int n = (int) Math.min(readBuffer.length, sz);
  1036. IO.readFully(cur, curBuffer, 0, n);
  1037. IO.readFully(pck, readBuffer, 0, n);
  1038. for (int i = 0; i < n; i++) {
  1039. if (curBuffer[i] != readBuffer[i]) {
  1040. throw new IOException(MessageFormat.format(
  1041. JGitText.get().collisionOn, obj.name()));
  1042. }
  1043. }
  1044. sz -= n;
  1045. }
  1046. }
  1047. } catch (MissingObjectException notLocal) {
  1048. // This is OK, we don't have a copy of the object locally
  1049. // but the API throws when we try to read it as usually it's
  1050. // an error to read something that doesn't exist.
  1051. }
  1052. }
  1053. private void checkObjectCollision(AnyObjectId obj, int type, byte[] data)
  1054. throws IOException {
  1055. try {
  1056. final ObjectLoader ldr = readCurs.open(obj, type);
  1057. final byte[] existingData = ldr.getCachedBytes(data.length);
  1058. if (!Arrays.equals(data, existingData)) {
  1059. throw new IOException(MessageFormat.format(
  1060. JGitText.get().collisionOn, obj.name()));
  1061. }
  1062. } catch (MissingObjectException notLocal) {
  1063. // This is OK, we don't have a copy of the object locally
  1064. // but the API throws when we try to read it as usually its
  1065. // an error to read something that doesn't exist.
  1066. }
  1067. }
  1068. /** @return current position of the input stream being parsed. */
  1069. private long streamPosition() {
  1070. return bBase + bOffset;
  1071. }
  1072. private ObjectTypeAndSize openDatabase(PackedObjectInfo obj,
  1073. ObjectTypeAndSize info) throws IOException {
  1074. bOffset = 0;
  1075. bAvail = 0;
  1076. return seekDatabase(obj, info);
  1077. }
  1078. private ObjectTypeAndSize openDatabase(UnresolvedDelta delta,
  1079. ObjectTypeAndSize info) throws IOException {
  1080. bOffset = 0;
  1081. bAvail = 0;
  1082. return seekDatabase(delta, info);
  1083. }
  1084. // Consume exactly one byte from the buffer and return it.
  1085. private int readFrom(Source src) throws IOException {
  1086. if (bAvail == 0)
  1087. fill(src, 1);
  1088. bAvail--;
  1089. return buf[bOffset++] & 0xff;
  1090. }
  1091. // Consume cnt bytes from the buffer.
  1092. void use(int cnt) {
  1093. bOffset += cnt;
  1094. bAvail -= cnt;
  1095. }
  1096. // Ensure at least need bytes are available in {@link #buf}.
  1097. int fill(Source src, int need) throws IOException {
  1098. while (bAvail < need) {
  1099. int next = bOffset + bAvail;
  1100. int free = buf.length - next;
  1101. if (free + bAvail < need) {
  1102. switch (src) {
  1103. case INPUT:
  1104. sync();
  1105. break;
  1106. case DATABASE:
  1107. if (bAvail > 0)
  1108. System.arraycopy(buf, bOffset, buf, 0, bAvail);
  1109. bOffset = 0;
  1110. break;
  1111. }
  1112. next = bAvail;
  1113. free = buf.length - next;
  1114. }
  1115. switch (src) {
  1116. case INPUT:
  1117. next = in.read(buf, next, free);
  1118. break;
  1119. case DATABASE:
  1120. next = readDatabase(buf, next, free);
  1121. break;
  1122. }
  1123. if (next <= 0)
  1124. throw new EOFException(
  1125. JGitText.get().packfileIsTruncatedNoParam);
  1126. bAvail += next;
  1127. }
  1128. return bOffset;
  1129. }
  1130. // Store consumed bytes in {@link #buf} up to {@link #bOffset}.
  1131. private void sync() throws IOException {
  1132. packDigest.update(buf, 0, bOffset);
  1133. onStoreStream(buf, 0, bOffset);
  1134. if (expectDataAfterPackFooter) {
  1135. if (bAvail > 0) {
  1136. in.reset();
  1137. IO.skipFully(in, bOffset);
  1138. bAvail = 0;
  1139. }
  1140. in.mark(buf.length);
  1141. } else if (bAvail > 0)
  1142. System.arraycopy(buf, bOffset, buf, 0, bAvail);
  1143. bBase += bOffset;
  1144. bOffset = 0;
  1145. }
  1146. /**
  1147. * Get a temporary byte array for use by the caller.
  1148. *
  1149. * @return a temporary byte array for use by the caller.
  1150. */
  1151. protected byte[] buffer() {
  1152. return tempBuffer;
  1153. }
  1154. /**
  1155. * Construct a PackedObjectInfo instance for this parser.
  1156. *
  1157. * @param id
  1158. * identity of the object to be tracked.
  1159. * @param delta
  1160. * if the object was previously an unresolved delta, this is the
  1161. * delta object that was tracking it. Otherwise null.
  1162. * @param deltaBase
  1163. * if the object was previously an unresolved delta, this is the
  1164. * ObjectId of the base of the delta. The base may be outside of
  1165. * the pack stream if the stream was a thin-pack.
  1166. * @return info object containing this object's data.
  1167. */
  1168. protected PackedObjectInfo newInfo(AnyObjectId id, UnresolvedDelta delta,
  1169. ObjectId deltaBase) {
  1170. PackedObjectInfo oe = new PackedObjectInfo(id);
  1171. if (delta != null)
  1172. oe.setCRC(delta.crc);
  1173. return oe;
  1174. }
  1175. /**
  1176. * Set the expected number of objects in the pack stream.
  1177. * <p>
  1178. * The object count in the pack header is not always correct for some Dfs
  1179. * pack files. e.g. INSERT pack always assume 1 object in the header since
  1180. * the actual object count is unknown when the pack is written.
  1181. * <p>
  1182. * If external implementation wants to overwrite the expectedObjectCount,
  1183. * they should call this method during {@link #onPackHeader(long)}.
  1184. *
  1185. * @param expectedObjectCount a long.
  1186. * @since 4.9
  1187. */
  1188. protected void setExpectedObjectCount(long expectedObjectCount) {
  1189. this.expectedObjectCount = expectedObjectCount;
  1190. }
  1191. /**
  1192. * Store bytes received from the raw stream.
  1193. * <p>
  1194. * This method is invoked during {@link #parse(ProgressMonitor)} as data is
  1195. * consumed from the incoming stream. Implementors may use this event to
  1196. * archive the raw incoming stream to the destination repository in large
  1197. * chunks, without paying attention to object boundaries.
  1198. * <p>
  1199. * The only component of the pack not supplied to this method is the last 20
  1200. * bytes of the pack that comprise the trailing SHA-1 checksum. Those are
  1201. * passed to {@link #onPackFooter(byte[])}.
  1202. *
  1203. * @param raw
  1204. * buffer to copy data out of.
  1205. * @param pos
  1206. * first offset within the buffer that is valid.
  1207. * @param len
  1208. * number of bytes in the buffer that are valid.
  1209. * @throws java.io.IOException
  1210. * the stream cannot be archived.
  1211. */
  1212. protected abstract void onStoreStream(byte[] raw, int pos, int len)
  1213. throws IOException;
  1214. /**
  1215. * Store (and/or checksum) an object header.
  1216. * <p>
  1217. * Invoked after any of the {@code onBegin()} events. The entire header is
  1218. * supplied in a single invocation, before any object data is supplied.
  1219. *
  1220. * @param src
  1221. * where the data came from
  1222. * @param raw
  1223. * buffer to read data from.
  1224. * @param pos
  1225. * first offset within buffer that is valid.
  1226. * @param len
  1227. * number of bytes in buffer that are valid.
  1228. * @throws java.io.IOException
  1229. * the stream cannot be archived.
  1230. */
  1231. protected abstract void onObjectHeader(Source src, byte[] raw, int pos,
  1232. int len) throws IOException;
  1233. /**
  1234. * Store (and/or checksum) a portion of an object's data.
  1235. * <p>
  1236. * This method may be invoked multiple times per object, depending on the
  1237. * size of the object, the size of the parser's internal read buffer, and
  1238. * the alignment of the object relative to the read buffer.
  1239. * <p>
  1240. * Invoked after {@link #onObjectHeader(Source, byte[], int, int)}.
  1241. *
  1242. * @param src
  1243. * where the data came from
  1244. * @param raw
  1245. * buffer to read data from.
  1246. * @param pos
  1247. * first offset within buffer that is valid.
  1248. * @param len
  1249. * number of bytes in buffer that are valid.
  1250. * @throws java.io.IOException
  1251. * the stream cannot be archived.
  1252. */
  1253. protected abstract void onObjectData(Source src, byte[] raw, int pos,
  1254. int len) throws IOException;
  1255. /**
  1256. * Invoked for commits, trees, tags, and small blobs.
  1257. *
  1258. * @param obj
  1259. * the object info, populated.
  1260. * @param typeCode
  1261. * the type of the object.
  1262. * @param data
  1263. * inflated data for the object.
  1264. * @throws java.io.IOException
  1265. * the object cannot be archived.
  1266. */
  1267. protected abstract void onInflatedObjectData(PackedObjectInfo obj,
  1268. int typeCode, byte[] data) throws IOException;
  1269. /**
  1270. * Provide the implementation with the original stream's pack header.
  1271. *
  1272. * @param objCnt
  1273. * number of objects expected in the stream.
  1274. * @throws java.io.IOException
  1275. * the implementation refuses to work with this many objects.
  1276. */
  1277. protected abstract void onPackHeader(long objCnt) throws IOException;
  1278. /**
  1279. * Provide the implementation with the original stream's pack footer.
  1280. *
  1281. * @param hash
  1282. * the trailing 20 bytes of the pack, this is a SHA-1 checksum of
  1283. * all of the pack data.
  1284. * @throws java.io.IOException
  1285. * the stream cannot be archived.
  1286. */
  1287. protected abstract void onPackFooter(byte[] hash) throws IOException;
  1288. /**
  1289. * Provide the implementation with a base that was outside of the pack.
  1290. * <p>
  1291. * This event only occurs on a thin pack for base objects that were outside
  1292. * of the pack and came from the local repository. Usually an implementation
  1293. * uses this event to compress the base and append it onto the end of the
  1294. * pack, so the pack stays self-contained.
  1295. *
  1296. * @param typeCode
  1297. * type of the base object.
  1298. * @param data
  1299. * complete content of the base object.
  1300. * @param info
  1301. * packed object information for this base. Implementors must
  1302. * populate the CRC and offset members if returning true.
  1303. * @return true if the {@code info} should be included in the object list
  1304. * returned by {@link #getSortedObjectList(Comparator)}, false if it
  1305. * should not be included.
  1306. * @throws java.io.IOException
  1307. * the base could not be included into the pack.
  1308. */
  1309. protected abstract boolean onAppendBase(int typeCode, byte[] data,
  1310. PackedObjectInfo info) throws IOException;
  1311. /**
  1312. * Event indicating a thin pack has been completely processed.
  1313. * <p>
  1314. * This event is invoked only if a thin pack has delta references to objects
  1315. * external from the pack. The event is called after all of those deltas
  1316. * have been resolved.
  1317. *
  1318. * @throws java.io.IOException
  1319. * the pack cannot be archived.
  1320. */
  1321. protected abstract void onEndThinPack() throws IOException;
  1322. /**
  1323. * Reposition the database to re-read a previously stored object.
  1324. * <p>
  1325. * If the database is computing CRC-32 checksums for object data, it should
  1326. * reset its internal CRC instance during this method call.
  1327. *
  1328. * @param obj
  1329. * the object position to begin reading from. This is from
  1330. * {@link #newInfo(AnyObjectId, UnresolvedDelta, ObjectId)}.
  1331. * @param info
  1332. * object to populate with type and size.
  1333. * @return the {@code info} object.
  1334. * @throws java.io.IOException
  1335. * the database cannot reposition to this location.
  1336. */
  1337. protected abstract ObjectTypeAndSize seekDatabase(PackedObjectInfo obj,
  1338. ObjectTypeAndSize info) throws IOException;
  1339. /**
  1340. * Reposition the database to re-read a previously stored object.
  1341. * <p>
  1342. * If the database is computing CRC-32 checksums for object data, it should
  1343. * reset its internal CRC instance during this method call.
  1344. *
  1345. * @param delta
  1346. * the object position to begin reading from. This is an instance
  1347. * previously returned by {@link #onEndDelta()}.
  1348. * @param info
  1349. * object to populate with type and size.
  1350. * @return the {@code info} object.
  1351. * @throws java.io.IOException
  1352. * the database cannot reposition to this location.
  1353. */
  1354. protected abstract ObjectTypeAndSize seekDatabase(UnresolvedDelta delta,
  1355. ObjectTypeAndSize info) throws IOException;
  1356. /**
  1357. * Read from the database's current position into the buffer.
  1358. *
  1359. * @param dst
  1360. * the buffer to copy read data into.
  1361. * @param pos
  1362. * position within {@code dst} to start copying data into.
  1363. * @param cnt
  1364. * ideal target number of bytes to read. Actual read length may
  1365. * be shorter.
  1366. * @return number of bytes stored.
  1367. * @throws java.io.IOException
  1368. * the database cannot be accessed.
  1369. */
  1370. protected abstract int readDatabase(byte[] dst, int pos, int cnt)
  1371. throws IOException;
  1372. /**
  1373. * Check the current CRC matches the expected value.
  1374. * <p>
  1375. * This method is invoked when an object is read back in from the database
  1376. * and its data is used during delta resolution. The CRC is validated after
  1377. * the object has been fully read, allowing the parser to verify there was
  1378. * no silent data corruption.
  1379. * <p>
  1380. * Implementations are free to ignore this check by always returning true if
  1381. * they are performing other data integrity validations at a lower level.
  1382. *
  1383. * @param oldCRC
  1384. * the prior CRC that was recorded during the first scan of the
  1385. * object from the pack stream.
  1386. * @return true if the CRC matches; false if it does not.
  1387. */
  1388. protected abstract boolean checkCRC(int oldCRC);
  1389. /**
  1390. * Event notifying the start of an object stored whole (not as a delta).
  1391. *
  1392. * @param streamPosition
  1393. * position of this object in the incoming stream.
  1394. * @param type
  1395. * type of the object; one of
  1396. * {@link org.eclipse.jgit.lib.Constants#OBJ_COMMIT},
  1397. * {@link org.eclipse.jgit.lib.Constants#OBJ_TREE},
  1398. * {@link org.eclipse.jgit.lib.Constants#OBJ_BLOB}, or
  1399. * {@link org.eclipse.jgit.lib.Constants#OBJ_TAG}.
  1400. * @param inflatedSize
  1401. * size of the object when fully inflated. The size stored within
  1402. * the pack may be larger or smaller, and is not yet known.
  1403. * @throws java.io.IOException
  1404. * the object cannot be recorded.
  1405. */
  1406. protected abstract void onBeginWholeObject(long streamPosition, int type,
  1407. long inflatedSize) throws IOException;
  1408. /**
  1409. * Event notifying the current object.
  1410. *
  1411. *@param info
  1412. * object information.
  1413. * @throws java.io.IOException
  1414. * the object cannot be recorded.
  1415. */
  1416. protected abstract void onEndWholeObject(PackedObjectInfo info)
  1417. throws IOException;
  1418. /**
  1419. * Event notifying start of a delta referencing its base by offset.
  1420. *
  1421. * @param deltaStreamPosition
  1422. * position of this object in the incoming stream.
  1423. * @param baseStreamPosition
  1424. * position of the base object in the incoming stream. The base
  1425. * must be before the delta, therefore {@code baseStreamPosition
  1426. * &lt; deltaStreamPosition}. This is <b>not</b> the position
  1427. * returned by a prior end object event.
  1428. * @param inflatedSize
  1429. * size of the delta when fully inflated. The size stored within
  1430. * the pack may be larger or smaller, and is not yet known.
  1431. * @throws java.io.IOException
  1432. * the object cannot be recorded.
  1433. */
  1434. protected abstract void onBeginOfsDelta(long deltaStreamPosition,
  1435. long baseStreamPosition, long inflatedSize) throws IOException;
  1436. /**
  1437. * Event notifying start of a delta referencing its base by ObjectId.
  1438. *
  1439. * @param deltaStreamPosition
  1440. * position of this object in the incoming stream.
  1441. * @param baseId
  1442. * name of the base object. This object may be later in the
  1443. * stream, or might not appear at all in the stream (in the case
  1444. * of a thin-pack).
  1445. * @param inflatedSize
  1446. * size of the delta when fully inflated. The size stored within
  1447. * the pack may be larger or smaller, and is not yet known.
  1448. * @throws java.io.IOException
  1449. * the object cannot be recorded.
  1450. */
  1451. protected abstract void onBeginRefDelta(long deltaStreamPosition,
  1452. AnyObjectId baseId, long inflatedSize) throws IOException;
  1453. /**
  1454. * Event notifying the current object.
  1455. *
  1456. *@return object information that must be populated with at least the
  1457. * offset.
  1458. * @throws java.io.IOException
  1459. * the object cannot be recorded.
  1460. */
  1461. protected UnresolvedDelta onEndDelta() throws IOException {
  1462. return new UnresolvedDelta();
  1463. }
  1464. /** Type and size information about an object in the database buffer. */
  1465. public static class ObjectTypeAndSize {
  1466. /** The type of the object. */
  1467. public int type;
  1468. /** The inflated size of the object. */
  1469. public long size;
  1470. }
  1471. private void inflateAndSkip(Source src, long inflatedSize)
  1472. throws IOException {
  1473. try (InputStream inf = inflate(src, inflatedSize)) {
  1474. IO.skipFully(inf, inflatedSize);
  1475. }
  1476. }
  1477. private byte[] inflateAndReturn(Source src, long inflatedSize)
  1478. throws IOException {
  1479. final byte[] dst = new byte[(int) inflatedSize];
  1480. try (InputStream inf = inflate(src, inflatedSize)) {
  1481. IO.readFully(inf, dst, 0, dst.length);
  1482. }
  1483. return dst;
  1484. }
  1485. private InputStream inflate(Source src, long inflatedSize)
  1486. throws IOException {
  1487. inflater.open(src, inflatedSize);
  1488. return inflater;
  1489. }
  1490. private static class DeltaChain extends ObjectIdOwnerMap.Entry {
  1491. UnresolvedDelta head;
  1492. DeltaChain(AnyObjectId id) {
  1493. super(id);
  1494. }
  1495. UnresolvedDelta remove() {
  1496. final UnresolvedDelta r = head;
  1497. if (r != null)
  1498. head = null;
  1499. return r;
  1500. }
  1501. void add(UnresolvedDelta d) {
  1502. d.next = head;
  1503. head = d;
  1504. }
  1505. }
  1506. /** Information about an unresolved delta in this pack stream. */
  1507. public static class UnresolvedDelta {
  1508. long position;
  1509. int crc;
  1510. UnresolvedDelta next;
  1511. /** @return offset within the input stream. */
  1512. public long getOffset() {
  1513. return position;
  1514. }
  1515. /** @return the CRC-32 checksum of the stored delta data. */
  1516. public int getCRC() {
  1517. return crc;
  1518. }
  1519. /**
  1520. * @param crc32
  1521. * the CRC-32 checksum of the stored delta data.
  1522. */
  1523. public void setCRC(int crc32) {
  1524. crc = crc32;
  1525. }
  1526. }
  1527. private static class DeltaVisit {
  1528. final UnresolvedDelta delta;
  1529. ObjectId id;
  1530. byte[] data;
  1531. DeltaVisit parent;
  1532. UnresolvedDelta nextChild;
  1533. DeltaVisit() {
  1534. this.delta = null; // At the root of the stack we have a base.
  1535. }
  1536. DeltaVisit(DeltaVisit parent) {
  1537. this.parent = parent;
  1538. this.delta = parent.nextChild;
  1539. parent.nextChild = delta.next;
  1540. }
  1541. DeltaVisit next() {
  1542. // If our parent has no more children, discard it.
  1543. if (parent != null && parent.nextChild == null) {
  1544. parent.data = null;
  1545. parent = parent.parent;
  1546. }
  1547. if (nextChild != null)
  1548. return new DeltaVisit(this);
  1549. // If we have no child ourselves, our parent must (if it exists),
  1550. // due to the discard rule above. With no parent, we are done.
  1551. if (parent != null)
  1552. return new DeltaVisit(parent);
  1553. return null;
  1554. }
  1555. }
  1556. private void addObjectAndTrack(PackedObjectInfo oe) {
  1557. entries[entryCount++] = oe;
  1558. if (needNewObjectIds())
  1559. newObjectIds.add(oe);
  1560. }
  1561. private class InflaterStream extends InputStream {
  1562. private final Inflater inf;
  1563. private final byte[] skipBuffer;
  1564. private Source src;
  1565. private long expectedSize;
  1566. private long actualSize;
  1567. private int p;
  1568. InflaterStream() {
  1569. inf = InflaterCache.get();
  1570. skipBuffer = new byte[512];
  1571. }
  1572. void release() {
  1573. inf.reset();
  1574. InflaterCache.release(inf);
  1575. }
  1576. void open(Source source, long inflatedSize) throws IOException {
  1577. src = source;
  1578. expectedSize = inflatedSize;
  1579. actualSize = 0;
  1580. p = fill(src, 1);
  1581. inf.setInput(buf, p, bAvail);
  1582. }
  1583. @Override
  1584. public long skip(long toSkip) throws IOException {
  1585. long n = 0;
  1586. while (n < toSkip) {
  1587. final int cnt = (int) Math.min(skipBuffer.length, toSkip - n);
  1588. final int r = read(skipBuffer, 0, cnt);
  1589. if (r <= 0)
  1590. break;
  1591. n += r;
  1592. }
  1593. return n;
  1594. }
  1595. @Override
  1596. public int read() throws IOException {
  1597. int n = read(skipBuffer, 0, 1);
  1598. return n == 1 ? skipBuffer[0] & 0xff : -1;
  1599. }
  1600. @Override
  1601. public int read(byte[] dst, int pos, int cnt) throws IOException {
  1602. try {
  1603. int n = 0;
  1604. while (n < cnt) {
  1605. int r = inf.inflate(dst, pos + n, cnt - n);
  1606. n += r;
  1607. if (inf.finished())
  1608. break;
  1609. if (inf.needsInput()) {
  1610. onObjectData(src, buf, p, bAvail);
  1611. use(bAvail);
  1612. p = fill(src, 1);
  1613. inf.setInput(buf, p, bAvail);
  1614. } else if (r == 0) {
  1615. throw new CorruptObjectException(MessageFormat.format(
  1616. JGitText.get().packfileCorruptionDetected,
  1617. JGitText.get().unknownZlibError));
  1618. }
  1619. }
  1620. actualSize += n;
  1621. return 0 < n ? n : -1;
  1622. } catch (DataFormatException dfe) {
  1623. throw new CorruptObjectException(MessageFormat.format(JGitText
  1624. .get().packfileCorruptionDetected, dfe.getMessage()));
  1625. }
  1626. }
  1627. @Override
  1628. public void close() throws IOException {
  1629. // We need to read here to enter the loop above and pump the
  1630. // trailing checksum into the Inflater. It should return -1 as the
  1631. // caller was supposed to consume all content.
  1632. //
  1633. if (read(skipBuffer) != -1 || actualSize != expectedSize) {
  1634. throw new CorruptObjectException(MessageFormat.format(JGitText
  1635. .get().packfileCorruptionDetected,
  1636. JGitText.get().wrongDecompressedLength));
  1637. }
  1638. int used = bAvail - inf.getRemaining();
  1639. if (0 < used) {
  1640. onObjectData(src, buf, p, used);
  1641. use(used);
  1642. }
  1643. inf.reset();
  1644. }
  1645. }
  1646. }