f50fee44b129d516772f00d7a1d00be2f3953e6f markd Thu Jul 2 11:50:46 2020 -0700 hgBlat appears to be working diff --git src/hg/hgBlat/hgBlat.c src/hg/hgBlat/hgBlat.c index cef6e0c..1188981 100644 --- src/hg/hgBlat/hgBlat.c +++ src/hg/hgBlat/hgBlat.c @@ -65,30 +65,31 @@ /* Information about hits on a genome assembly */ { struct genomeHits *next; char *db; /* Database name. */ char *genome; /* Genome name. */ int seqNumber; /* Submission order */ char *faName; /* fasta name */ char *dna; /* query dna */ int dnaSize; /* query dna size */ int sd; /* Connection */ char *type; /* query type = query, protQuery, transQuery */ char *xType; /* query type = dna, prot, rnax, dnax */ boolean queryRC; /* is the query reverse-complemented */ boolean complex; /* is the query complex */ boolean isProt; /* is the protein query */ + char *dynGenome; /* genome for dynamic server */ int maxGeneHits; /* Highest gene hit-count */ char *maxGeneChrom; /* Target Chrom for gene with max gene hits */ int maxGeneChromSize; /* Target Chrom Size for only prot, rnax, dnax */ int maxGeneTStart; /* Target Start Coordinate for gene with max hits */ int maxGeneTEnd; /* Target End Coordinate for gene with max hits*/ int maxGeneExons; /* Number of Exons in gene with max hits */ char maxGeneStrand[3]; /* + or - or ++ +- -+ -- Strand for gene with max hits */ char maxGeneTStrand;/* + or - TStrand for gene with max hits */ boolean done; /* Did the job get to finish */ boolean error; /* Some error happened */ char *networkErrMsg; /* Network layer error message */ struct dyString *dbg; /* Output debugging info */ struct gfResult *gfList; /* List of gfResult records */ boolean hide; /* To not show both strands, suppress the weaker-scoring one */ @@ -837,48 +838,49 @@ { safef(shortName, sizeof shortName, "blat %s+%d", names->name, count - 1); safef(description, sizeof description, "blat on %d queries (%s, %s)", count, names->name, names->next->name); } else { safef(shortName, sizeof shortName, "blat %s+%d", names->name, count - 1); safef(description, sizeof description, "blat on %d queries (%s, %s, ...)", count, names->name, names->next->name); } *pName = makeNameUnique(shortName, database, cart); *pDescription = cloneString(description); } void queryServerSetup(int conn, char *db, struct dnaSeq *seq, char *type, char *xType, - boolean complex, boolean isProt, boolean queryRC, int seqNumber) + boolean complex, boolean isProt, boolean queryRC, int seqNumber, char *dynGenome) /* Setup query to server. * queryRC is true when the query has been reverse-complemented */ { struct genomeHits *gH; AllocVar(gH); gH->db = cloneString(db); gH->genome = cloneString(hGenome(db)); gH->seqNumber = seqNumber; gH->faName = cloneString(seq->name); gH->dna = cloneString(seq->dna); gH->dnaSize = seq->size; gH->type = cloneString(type); gH->xType = cloneString(xType); gH->queryRC = queryRC; gH->complex = complex; gH->isProt = isProt; +gH->dynGenome = dynGenome; gH->sd = conn; if (gH->sd == -1) { gH->error = TRUE; gH->networkErrMsg = "Connection to gfServer failed."; } gH->dbg = dyStringNew(256); slAddHead(&pfdList, gH); } void findBestGene(struct genomeHits *gH, int queryFrame) /* Find best gene-like object with multiple linked-features. * Remember chrom start end of best gene found and total hits in the gene. * Should sort the gfResults by tStrand, chrom, tStart. * Filters on queryFrame */ @@ -976,31 +978,34 @@ gfR->qEnd = gfR->qEnd * qFactor + gfR->qFrame; gfR->tStart = gfR->tStart * tFactor + gfR->tFrame; gfR->tEnd = gfR->tEnd * tFactor + gfR->tFrame; } } void queryServerFinish(struct genomeHits *gH) /* Report results from gfServer. */ { char buf[256]; int matchCount = 0; dyStringPrintf(gH->dbg,"query strand %s qsize %d<br>\n", gH->queryRC ? "-" : "+", gH->dnaSize); /* Put together query command. */ +if (gH->dynGenome == NULL) safef(buf, sizeof buf, "%s%s %d", gfSignature(), gH->type, gH->dnaSize); +else + safef(buf, sizeof buf, "%s%s %d %s", gfSignature(), gH->type, gH->dnaSize, gH->dynGenome); mustWriteFd(gH->sd, buf, strlen(buf)); if (read(gH->sd, buf, 1) < 0) errAbort("queryServerFinish: read failed: %s", strerror(errno)); if (buf[0] != 'Y') errAbort("Expecting 'Y' from server, got %c", buf[0]); mustWriteFd(gH->sd, gH->dna, gH->dnaSize); if (gH->complex) { char *s = netRecieveString(gH->sd, buf); if (!s) errAbort("expected response from gfServer with tileSize"); dyStringPrintf(gH->dbg,"%s<br>\n", s); // from server: tileSize 4 } @@ -1299,31 +1304,35 @@ } int findGenomeParams(struct serverTable *serve) /* Send status message to server arnd report result. * Get tileSize stepSize and minMatch. */ { char buf[256]; int sd = 0; int ret = 0; /* Put together command. */ sd = gfConnectEx(serve->host, serve->port); +if (serve->dynGenome == NULL) sprintf(buf, "%sstatus", gfSignature()); +else + sprintf(buf, "%s%s %s", gfSignature(), (serve->isTrans ? "transInfo" : "untransInfo"), + serve->dynGenome); mustWriteFd(sd, buf, strlen(buf)); for (;;) { if (netGetString(sd, buf) == NULL) { warn("Error reading status information from %s:%s",serve->host, serve->port); ret = -1; break; } if (sameString(buf, "end")) break; else { if (startsWith("tileSize ", buf)) @@ -1332,64 +1341,30 @@ } if (startsWith("stepSize ", buf)) { serve->stepSize = atoi(buf+strlen("stepSize ")); } if (startsWith("minMatch ", buf)) { serve->minMatch = atoi(buf+strlen("minMatch ")); } } } close(sd); return(ret); } -int findGenomeParamsDynamic(struct serverTable *serve) -/* read genome params from dynamic server, which follow "end" in response trailer. - */ -{ -char buf[256]; -for (;;) - { - if (netGetString(STDIN_FILENO, buf) == NULL) - { - warn("Error reading status information from dynamic server"); - return -1; - } - if (sameString(buf, "trailerEnd")) - return 0; - else - { - if (startsWith("tileSize ", buf)) - { - serve->tileSize = atoi(buf+strlen("tileSize ")); - } - if (startsWith("stepSize ", buf)) - { - serve->stepSize = atoi(buf+strlen("stepSize ")); - } - if (startsWith("minMatch ", buf)) - { - serve->minMatch = atoi(buf+strlen("minMatch ")); - } - } - } -} - - - void blatSeq(char *userSeq, char *organism, char *database, int dbCount) /* Blat sequence user pasted in. */ { FILE *f; struct dnaSeq *seqList = NULL, *seq; struct tempName pslTn, faTn; int maxSingleSize, maxTotalSize, maxSeqCount; char *genome, *db; char *type = cgiString("type"); char *seqLetters = cloneString(userSeq); struct serverTable *serve; int conn; int oneSize, totalSize = 0, seqCount = 0; boolean isTx = FALSE; boolean isTxTx = FALSE; @@ -1564,33 +1539,30 @@ // read genome size if (trackHubDatabase(database)) { genomeSize = findGenomeSizeFromHub(database); } else { genomeSize = findGenomeSize(database); } minMatchShown = findMinMatch(genomeSize, qType == gftProt); #endif if (allResults) minMatchShown = 0; // read tileZize stepSize minMatch from server status - if (serve->dynGenome != NULL) - findGenomeParamsDynamic(serve); - else findGenomeParams(serve); int minLucky = (serve->minMatch * serve->stepSize + (serve->tileSize - serve->stepSize)) * xlat; minSuggested = max(minMatchShown,minLucky); } int seqNumber = 0; /* Loop through each sequence. */ for (seq = seqList; seq != NULL; seq = seq->next) { printf(" "); fflush(stdout); /* prevent apache cgi timeout by outputting something */ oneSize = realSeqSize(seq, !isTx); // Impose half the usual bot delay per sequence @@ -1622,65 +1594,70 @@ if (totalSize > maxTotalSize) { warn("Sequence %s would take us over the %d letter limit, stopping here.", seq->name, maxTotalSize); break; } conn = gfConnectEx(serve->host, serve->port); if (isTx) { gvo->reportTargetStrand = TRUE; if (isTxTx) { if (allGenomes) - queryServerSetup(conn, db, seq, "transQuery", xType, TRUE, FALSE, FALSE, seqNumber); + queryServerSetup(conn, db, seq, "transQuery", xType, TRUE, FALSE, FALSE, seqNumber, + serve->dynGenome); else gfAlignTransTrans(&conn, serve->nibDir, seq, FALSE, 5, tFileCache, gvo, !txTxBoth, serve->dynGenome); if (txTxBoth) { reverseComplement(seq->dna, seq->size); conn = gfConnectEx(serve->host, serve->port); if (allGenomes) - queryServerSetup(conn, db, seq, "transQuery", xType, TRUE, FALSE, TRUE, seqNumber); + queryServerSetup(conn, db, seq, "transQuery", xType, TRUE, FALSE, TRUE, seqNumber, + serve->dynGenome); else gfAlignTransTrans(&conn, serve->nibDir, seq, TRUE, 5, tFileCache, gvo, FALSE, serve->dynGenome); } } else { if (allGenomes) - queryServerSetup(conn, db, seq, "protQuery", xType, TRUE, TRUE, FALSE, seqNumber); + queryServerSetup(conn, db, seq, "protQuery", xType, TRUE, TRUE, FALSE, seqNumber, + serve->dynGenome); else gfAlignTrans(&conn, serve->nibDir, seq, 5, tFileCache, gvo, serve->dynGenome); } } else { if (allGenomes) - queryServerSetup(conn, db, seq, "query", xType, FALSE, FALSE, FALSE, seqNumber); + queryServerSetup(conn, db, seq, "query", xType, FALSE, FALSE, FALSE, seqNumber, + serve->dynGenome); else gfAlignStrand(&conn, serve->nibDir, seq, FALSE, minMatchShown, tFileCache, gvo, serve->dynGenome); reverseComplement(seq->dna, seq->size); conn = gfConnectEx(serve->host, serve->port); if (allGenomes) - queryServerSetup(conn, db, seq, "query", xType, FALSE, FALSE, TRUE, seqNumber); + queryServerSetup(conn, db, seq, "query", xType, FALSE, FALSE, TRUE, seqNumber, + serve->dynGenome); else gfAlignStrand(&conn, serve->nibDir, seq, TRUE, minMatchShown, tFileCache, gvo, serve->dynGenome); } gfOutputQuery(gvo, f); ++seqNumber; } carefulClose(&f); if (!allGenomes) { showAliPlaces(pslTn.forCgi, faTn.forCgi, NULL, serve->db, qType, tType, organism, feelingLucky); }