| Index: branches/ariel/xmldumps-backup/worker.py |
| — | — | @@ -609,7 +609,7 @@ |
| 610 | 610 | "metahistory7zdump", |
| 611 | 611 | "All pages with complete edit history (.7z)", |
| 612 | 612 | "These dumps can be *very* large, uncompressing up to 100 times the archive download size. " + |
| 613 | | - "Suitable for archival and statistical use, most mirror sites won't want or need this.", self.findItemByName('metahistorybz2dump'), self.wiki, self._getChunkToDo("metahistory7zdump"), self.chunkInfo.getPagesPerChunkHistory(), self.checkpointFile)) |
| | 613 | + "Suitable for archival and statistical use, most mirror sites won't want or need this.", self.findItemByName('metahistorybz2dump'), self.wiki, self._getChunkToDo("metahistory7zdump"), self.chunkInfo.getPagesPerChunkHistory(), checkpoints, self.checkpointFile)) |
| 614 | 614 | if (self.chunkInfo.chunksEnabled() and self.chunkInfo.recombineHistory()): |
| 615 | 615 | self.dumpItems.append( |
| 616 | 616 | RecombineXmlRecompressDump("metahistory7zdumprecombine", |
| — | — | @@ -1814,9 +1814,7 @@ |
| 1815 | 1815 | # of that very file. meh. how likely is it that we |
| 1816 | 1816 | # have one? these files are time based and the start/end pageids |
| 1817 | 1817 | # are going to fluctuate. whatever |
| 1818 | | - cf = DumpFilename(self.wiki) |
| 1819 | | - cf.newFromFilename(item.checkpointFile) |
| 1820 | | - checkpoint = cf.checkpoint |
| | 1818 | + checkpoint = item.checkpointFile.checkpoint |
| 1821 | 1819 | |
| 1822 | 1820 | for d in dumpNames: |
| 1823 | 1821 | self.symLinks.removeSymLinksFromOldRuns(self.wiki.date, d, chunk, checkpoint ) |
| — | — | @@ -3135,7 +3133,7 @@ |
| 3136 | 3134 | class XmlRecompressDump(Dump): |
| 3137 | 3135 | """Take a .bz2 and recompress it as 7-Zip.""" |
| 3138 | 3136 | |
| 3139 | | - def __init__(self, subset, name, desc, detail, itemForRecompression, wiki, chunkToDo, chunks = False, checkpoints = False): |
| | 3137 | + def __init__(self, subset, name, desc, detail, itemForRecompression, wiki, chunkToDo, chunks = False, checkpoints = False, checkpointFile = None): |
| 3140 | 3138 | self._subset = subset |
| 3141 | 3139 | self._detail = detail |
| 3142 | 3140 | self._chunks = chunks |
| — | — | @@ -3146,6 +3144,7 @@ |
| 3147 | 3145 | self.itemForRecompression = itemForRecompression |
| 3148 | 3146 | if checkpoints: |
| 3149 | 3147 | self._checkpointsEnabled = True |
| | 3148 | + self.checkpointFile = checkpointFile |
| 3150 | 3149 | Dump.__init__(self, name, desc) |
| 3151 | 3150 | |
| 3152 | 3151 | def getDumpName(self): |
| — | — | @@ -3182,7 +3181,11 @@ |
| 3183 | 3182 | commands = [] |
| 3184 | 3183 | # Remove prior 7zip attempts; 7zip will try to append to an existing archive |
| 3185 | 3184 | self.cleanupOldFiles(runner.dumpDir) |
| 3186 | | - if self._chunksEnabled and not self._chunkToDo: |
| | 3185 | + if self.checkpointFile: |
| | 3186 | + outputFile = DumpFilename(self.wiki, None, self.checkpointFile.dumpName, self.checkpointFile.fileType, self.fileExt, self.checkpointFile.chunk, self.checkpointFile.checkpoint) |
| | 3187 | + series = self.buildCommand(runner, [ outputFile ]) |
| | 3188 | + commands.append(series) |
| | 3189 | + elif self._chunksEnabled and not self._chunkToDo: |
| 3187 | 3190 | # must set up each parallel job separately, they may have checkpoint files that |
| 3188 | 3191 | # need to be processed in series, it's a special case |
| 3189 | 3192 | for i in range(1, len(self._chunks)+1): |