int main(int argc, char *argv[]) { /* * Open the database. * Construct a flower. * Construct an event tree representing the species tree. * For each sequence contruct two ends each containing an cap. * Make a file for the sequence. * Link the two caps. * Finish! */ int64_t key, j; Group *group; Flower_EndIterator *endIterator; End *end; bool makeEventHeadersAlphaNumeric = 0; /* * Arguments/options */ char * logLevelString = NULL; char * speciesTree = NULL; char * outgroupEvents = NULL; /////////////////////////////////////////////////////////////////////////// // (0) Parse the inputs handed by genomeCactus.py / setup stuff. /////////////////////////////////////////////////////////////////////////// while (1) { static struct option long_options[] = { { "logLevel", required_argument, 0, 'a' }, { "cactusDisk", required_argument, 0, 'b' }, { "speciesTree", required_argument, 0, 'f' }, { "outgroupEvents", required_argument, 0, 'g' }, { "help", no_argument, 0, 'h' }, { "makeEventHeadersAlphaNumeric", no_argument, 0, 'i' }, { 0, 0, 0, 0 } }; int option_index = 0; key = getopt_long(argc, argv, "a:b:f:hg:i", long_options, &option_index); if (key == -1) { break; } switch (key) { case 'a': logLevelString = optarg; break; case 'b': cactusDiskDatabaseString = optarg; break; case 'f': speciesTree = optarg; break; case 'g': outgroupEvents = optarg; break; case 'h': usage(); return 0; case 'i': makeEventHeadersAlphaNumeric = 1; break; default: usage(); return 1; } } /////////////////////////////////////////////////////////////////////////// // (0) Check the inputs. /////////////////////////////////////////////////////////////////////////// //assert(logLevelString == NULL || strcmp(logLevelString, "CRITICAL") == 0 || strcmp(logLevelString, "INFO") == 0 || strcmp(logLevelString, "DEBUG") == 0); assert(cactusDiskDatabaseString != NULL); assert(speciesTree != NULL); ////////////////////////////////////////////// //Set up logging ////////////////////////////////////////////// st_setLogLevelFromString(logLevelString); ////////////////////////////////////////////// //Log (some of) the inputs ////////////////////////////////////////////// st_logInfo("Flower disk name : %s\n", cactusDiskDatabaseString); for (j = optind; j < argc; j++) { st_logInfo("Sequence file/directory %s\n", argv[j]); } ////////////////////////////////////////////// //Load the database ////////////////////////////////////////////// stKVDatabaseConf *kvDatabaseConf = kvDatabaseConf = stKVDatabaseConf_constructFromString(cactusDiskDatabaseString); if (stKVDatabaseConf_getType(kvDatabaseConf) == stKVDatabaseTypeTokyoCabinet || stKVDatabaseConf_getType(kvDatabaseConf) == stKVDatabaseTypeKyotoTycoon) { assert(stKVDatabaseConf_getDir(kvDatabaseConf) != NULL); cactusDisk = cactusDisk_construct2(kvDatabaseConf, "cactusSequences"); } else { cactusDisk = cactusDisk_construct(kvDatabaseConf, 1); } st_logInfo("Set up the flower disk\n"); ////////////////////////////////////////////// //Construct the flower ////////////////////////////////////////////// if (cactusDisk_getFlower(cactusDisk, 0) != NULL) { cactusDisk_destruct(cactusDisk); st_logInfo("The first flower already exists\n"); return 0; } flower = flower_construct2(0, cactusDisk); assert(flower_getName(flower) == 0); st_logInfo("Constructed the flower\n"); ////////////////////////////////////////////// //Construct the event tree ////////////////////////////////////////////// st_logInfo("Going to build the event tree with newick string: %s\n", speciesTree); stTree *tree = stTree_parseNewickString(speciesTree); st_logInfo("Parsed the tree\n"); if (makeEventHeadersAlphaNumeric) { makeEventHeadersAlphaNumericFn(tree); } stTree_setBranchLength(tree, INT64_MAX); checkBranchLengthsAreDefined(tree); eventTree = eventTree_construct2(flower); //creates the event tree and the root even totalEventNumber = 1; st_logInfo("Constructed the basic event tree\n"); // Construct a set of outgroup names so that ancestral outgroups // get recognized. stSet *outgroupNameSet = stSet_construct3(stHash_stringKey, stHash_stringEqualKey, free); if(outgroupEvents != NULL) { stList *outgroupNames = stString_split(outgroupEvents); for(int64_t i = 0; i < stList_length(outgroupNames); i++) { char *outgroupName = stList_get(outgroupNames, i); stSet_insert(outgroupNameSet, stString_copy(outgroupName)); } stList_destruct(outgroupNames); } //now traverse the tree j = optind; assignEventsAndSequences(eventTree_getRootEvent(eventTree), tree, outgroupNameSet, argv, &j); char *eventTreeString = eventTree_makeNewickString(eventTree); st_logInfo( "Constructed the initial flower with %" PRIi64 " sequences and %" PRIi64 " events with string: %s\n", totalSequenceNumber, totalEventNumber, eventTreeString); assert(event_getSubTreeBranchLength(eventTree_getRootEvent(eventTree)) >= 0.0); free(eventTreeString); //assert(0); ////////////////////////////////////////////// //Label any outgroup events. ////////////////////////////////////////////// if (outgroupEvents != NULL) { stList *outgroupEventsList = stString_split(outgroupEvents); for (int64_t i = 0; i < stList_length(outgroupEventsList); i++) { char *outgroupEvent = makeEventHeadersAlphaNumeric ? makeAlphaNumeric(stList_get(outgroupEventsList, i)) : stString_copy(stList_get(outgroupEventsList, i)); Event *event = eventTree_getEventByHeader(eventTree, outgroupEvent); if (event == NULL) { st_errAbort("Got an outgroup string that does not match an event, outgroup string %s", outgroupEvent); } assert(!event_isOutgroup(event)); event_setOutgroupStatus(event, 1); assert(event_isOutgroup(event)); free(outgroupEvent); } stList_destruct(outgroupEventsList); } ////////////////////////////////////////////// //Construct the terminal group. ////////////////////////////////////////////// if (flower_getEndNumber(flower) > 0) { group = group_construct2(flower); endIterator = flower_getEndIterator(flower); while ((end = flower_getNextEnd(endIterator)) != NULL) { end_setGroup(end, group); } flower_destructEndIterator(endIterator); assert(group_isLeaf(group)); // Create a one link chain if there is only one pair of attached ends.. group_constructChainForLink(group); assert(!flower_builtBlocks(flower)); } else { flower_setBuiltBlocks(flower, 1); } /////////////////////////////////////////////////////////////////////////// // Write the flower to disk. /////////////////////////////////////////////////////////////////////////// //flower_check(flower); cactusDisk_write(cactusDisk); st_logInfo("Updated the flower on disk\n"); /////////////////////////////////////////////////////////////////////////// // Cleanup. /////////////////////////////////////////////////////////////////////////// cactusDisk_destruct(cactusDisk); return 0; //Exit without clean up is quicker, enable cleanup when doing memory leak detection. stSet_destruct(outgroupNameSet); stTree_destruct(tree); stKVDatabaseConf_destruct(kvDatabaseConf); return 0; }
int main(int argc, char *argv[]) { /* * Script for adding a reference genome to a flower. */ /* * Arguments/options */ char * logLevelString = NULL; char * cactusDiskDatabaseString = NULL; char *referenceEventString = (char *) cactusMisc_getDefaultReferenceEventHeader(); char *outputFile = NULL; Name flowerName = NULL_NAME; /////////////////////////////////////////////////////////////////////////// // (0) Parse the inputs handed by genomeCactus.py / setup stuff. /////////////////////////////////////////////////////////////////////////// while (1) { static struct option long_options[] = { { "logLevel", required_argument, 0, 'a' }, { "cactusDisk", required_argument, 0, 'c' }, { "flowerName", required_argument, 0, 'd' }, { "referenceEventString", required_argument, 0, 'g' }, { "help", no_argument, 0, 'h' }, { "outputFile", required_argument, 0, 'k' }, { 0, 0, 0, 0 } }; int option_index = 0; int key = getopt_long(argc, argv, "a:c:d:g:hk:", long_options, &option_index); if (key == -1) { break; } switch (key) { case 'a': logLevelString = stString_copy(optarg); break; case 'c': cactusDiskDatabaseString = stString_copy(optarg); break; case 'd': flowerName = cactusMisc_stringToName(optarg); break; case 'g': referenceEventString = stString_copy(optarg); break; case 'h': usage(); return 0; case 'k': outputFile = stString_copy(optarg); break; default: usage(); return 1; } } /////////////////////////////////////////////////////////////////////////// // (0) Check the inputs. /////////////////////////////////////////////////////////////////////////// assert(cactusDiskDatabaseString != NULL); ////////////////////////////////////////////// //Set up logging ////////////////////////////////////////////// st_setLogLevelFromString(logLevelString); ////////////////////////////////////////////// //Load the database ////////////////////////////////////////////// stKVDatabaseConf *kvDatabaseConf = stKVDatabaseConf_constructFromString( cactusDiskDatabaseString); CactusDisk *cactusDisk = cactusDisk_construct(kvDatabaseConf, 0); stKVDatabaseConf_destruct(kvDatabaseConf); st_logInfo("Set up the flower disk\n"); /////////////////////////////////////////////////////////////////////////// // Get the set of flowers to manipulate /////////////////////////////////////////////////////////////////////////// Flower *flower = cactusDisk_getFlower(cactusDisk, flowerName); /////////////////////////////////////////////////////////////////////////// // Get the reference event name /////////////////////////////////////////////////////////////////////////// Event *referenceEvent = eventTree_getEventByHeader( flower_getEventTree(flower), referenceEventString); assert(referenceEvent != NULL); Name referenceEventName = event_getName(referenceEvent); /////////////////////////////////////////////////////////////////////////// // Now process each flower in turn. /////////////////////////////////////////////////////////////////////////// if(outputFile == NULL) { st_errAbort("No output file specified\n"); } FILE *fileHandle = fopen(outputFile, "w"); printFastaSequences(flower, fileHandle, referenceEventName); if(fileHandle != NULL) { fclose(fileHandle); } /////////////////////////////////////////////////////////////////////////// //Clean up memory /////////////////////////////////////////////////////////////////////////// cactusDisk_destruct(cactusDisk); //return 0; //Exit without clean up is quicker, enable cleanup when doing memory leak detection. free(cactusDiskDatabaseString); free(referenceEventString); free(logLevelString); st_logInfo("Cleaned stuff up and am finished\n"); //while(1); return 0; }
int main(int argc, char *argv[]) { /* * Script for adding a reference genome to a flower. */ /* * Arguments/options */ char * logLevelString = NULL; char * cactusDiskDatabaseString = NULL; char * secondaryDatabaseString = NULL; char *referenceEventString = (char *) cactusMisc_getDefaultReferenceEventHeader(); bool bottomUpPhase = 0; /////////////////////////////////////////////////////////////////////////// // (0) Parse the inputs handed by genomeCactus.py / setup stuff. /////////////////////////////////////////////////////////////////////////// while (1) { static struct option long_options[] = { { "logLevel", required_argument, 0, 'a' }, { "cactusDisk", required_argument, 0, 'b' }, { "secondaryDisk", required_argument, 0, 'd' }, { "referenceEventString", required_argument, 0, 'g' }, { "help", no_argument, 0, 'h' }, { "bottomUpPhase", no_argument, 0, 'j' }, { 0, 0, 0, 0 } }; int option_index = 0; int key = getopt_long(argc, argv, "a:b:c:d:e:g:hi:j", long_options, &option_index); if (key == -1) { break; } switch (key) { case 'a': logLevelString = stString_copy(optarg); break; case 'b': cactusDiskDatabaseString = stString_copy(optarg); break; case 'd': secondaryDatabaseString = stString_copy(optarg); break; case 'g': referenceEventString = stString_copy(optarg); break; case 'h': usage(); return 0; case 'j': bottomUpPhase = 1; break; default: usage(); return 1; } } /////////////////////////////////////////////////////////////////////////// // (0) Check the inputs. /////////////////////////////////////////////////////////////////////////// assert(cactusDiskDatabaseString != NULL); ////////////////////////////////////////////// //Set up logging ////////////////////////////////////////////// st_setLogLevelFromString(logLevelString); ////////////////////////////////////////////// //Load the database ////////////////////////////////////////////// st_logInfo("referenceEventString = %s\n", referenceEventString); st_logInfo("bottomUpPhase = %i\n", bottomUpPhase); stKVDatabaseConf *kvDatabaseConf = stKVDatabaseConf_constructFromString(cactusDiskDatabaseString); CactusDisk *cactusDisk = cactusDisk_construct(kvDatabaseConf, false, true); stKVDatabaseConf_destruct(kvDatabaseConf); st_logInfo("Set up the flower disk\n"); stKVDatabase *sequenceDatabase = NULL; if (secondaryDatabaseString != NULL) { kvDatabaseConf = stKVDatabaseConf_constructFromString(secondaryDatabaseString); sequenceDatabase = stKVDatabase_construct(kvDatabaseConf, 0); stKVDatabaseConf_destruct(kvDatabaseConf); } FlowerStream *flowerStream = flowerWriter_getFlowerStream(cactusDisk, stdin); Flower *flower; while ((flower = flowerStream_getNext(flowerStream)) != NULL) { st_logDebug("Processing flower %" PRIi64 "\n", flower_getName(flower)); /////////////////////////////////////////////////////////////////////////// // Get the appropriate event names /////////////////////////////////////////////////////////////////////////// st_logInfo("%s\n", eventTree_makeNewickString(flower_getEventTree(flower))); Event *referenceEvent = eventTree_getEventByHeader(flower_getEventTree(flower), referenceEventString); if (referenceEvent == NULL) { st_errAbort("Reference event %s not found in tree. Check your " "--referenceEventString option", referenceEventString); } Name referenceEventName = event_getName(referenceEvent); /////////////////////////////////////////////////////////////////////////// // Now do bottom up or top down, depending /////////////////////////////////////////////////////////////////////////// stList *flowers = stList_construct(); stList_append(flowers, flower); preCacheNestedFlowers(cactusDisk, flowers); if (bottomUpPhase) { assert(sequenceDatabase != NULL); cactusDisk_preCacheSegmentStrings(cactusDisk, flowers); bottomUp(flowers, sequenceDatabase, referenceEventName, !flower_hasParentGroup(flower), generateJukesCantorMatrix); // Unload the nested flowers to save memory. They haven't // been changed, so we don't write them to the cactus // disk. Flower_GroupIterator *groupIt = flower_getGroupIterator(flower); Group *group; while ((group = flower_getNextGroup(groupIt)) != NULL) { if (!group_isLeaf(group)) { flower_unload(group_getNestedFlower(group)); } } flower_destructGroupIterator(groupIt); assert(!flower_isParentLoaded(flower)); // Write this flower to disk. cactusDisk_addUpdateRequest(cactusDisk, flower); } else { topDown(flower, referenceEventName); // We've changed the nested flowers, but not this // flower. We write the nested flowers to disk, then // unload them to save memory. This flower will be // unloaded by the flower-stream code. Flower_GroupIterator *groupIt = flower_getGroupIterator(flower); Group *group; while ((group = flower_getNextGroup(groupIt)) != NULL) { if (!group_isLeaf(group)) { cactusDisk_addUpdateRequest(cactusDisk, group_getNestedFlower(group)); flower_unload(group_getNestedFlower(group)); } } flower_destructGroupIterator(groupIt); } stList_destruct(flowers); } /////////////////////////////////////////////////////////////////////////// // Write the flower(s) back to disk. /////////////////////////////////////////////////////////////////////////// cactusDisk_write(cactusDisk); st_logInfo("Updated the flower on disk\n"); /////////////////////////////////////////////////////////////////////////// //Clean up. /////////////////////////////////////////////////////////////////////////// if (sequenceDatabase != NULL) { stKVDatabase_destruct(sequenceDatabase); } cactusDisk_destruct(cactusDisk); return 0; //Exit without clean up is quicker, enable cleanup when doing memory leak detection. free(cactusDiskDatabaseString); free(referenceEventString); free(logLevelString); st_logInfo("Cleaned stuff up and am finished\n"); return 0; }