void doMiddle()
/* Write middle part of .html. */
{
DNA *targetDna;
char *chrom;
int tStart, tEnd;
struct xaAli *xa;
int bothStart, bothEnd;
char cbCosmidName[256];
char *s;
/* Get input variables from CGI. */
char *qOrganism = cgiString("qOrganism");
char *tOrganism = cgiString("tOrganism");
char *query = cgiString("query");
char *target = cgiString("target");
char *strandString = cgiString("strand");
char strand = strandString[0];
boolean showSym = cgiBoolean("symbols");
boolean gotClickPos = cgiVarExists("clickPos");
double clickPos;
if (gotClickPos) clickPos = cgiDouble("clickPos");
strcpy(cbCosmidName, query);
if ((s = strrchr(cbCosmidName, '.')) != NULL)
*s = 0;
/* Get xaAli. */
xa = getOneXaAli(qOrganism, query);
printf("<H2>Alignment of <I>C. briggsae</I> %s:%d-%d and <I>C. elegans</I> %s</H2>\n",
cbCosmidName, xa->qStart, xa->qEnd, target);
htmlParagraph("<I>C. briggsae</I> appears on top. Coding regions in <I>C. elegans</I> are in upper case.");
/* Get display window. */
if (!wormParseChromRange(target, &chrom, &tStart, &tEnd))
errAbort("Target %s isn't formatted correctly", target);
/* Figure out intersection of display window and xeno-alignment */
bothStart = max(xa->tStart, tStart);
bothEnd = min(xa->tEnd, tEnd);
/* Get upper-cased-exon target DNA. */
targetDna = wormChromPartExonsUpper(chrom, bothStart, bothEnd - bothStart);
upcCorresponding(targetDna, bothEnd - bothStart, xa->tSym, bothStart - xa->tStart);
printf("<TT><PRE>");
showTargetRange(xa, bothStart - xa->tStart, bothEnd-bothStart, strand, showSym);
printf("</TT></PRE>");
}
void doFuzzyFind()
/* Do fuzzy-finder alignment. */
{
char *bacAcc = cgiString("bacAcc");
char *repeatMask = cgiString("repeatMask");
int contig = cgiInt("contig");
int qStart = cgiInt("qStart");
int qSize = cgiInt("qSize");
int tStart = cgiInt("tStart");
int tSize = cgiInt("tSize");
showDetailedMatch(bacAcc, contig, qStart, qSize, tStart, tSize, repeatMask);
}
void checkYbr(char *agpFile, char *faFile, char *mdFile)
/* checkYbr - Check NCBI assembly (aka Yellow Brick Road). */
{
struct hash *hash = newHash(0);
struct contig *contigList = NULL, *contig;
int problemCount = 0;
contigList = contigsFromAgp(agpFile, hash);
printf("Read %d contigs from %s\n", slCount(contigList), agpFile);
if (cgiVarExists("checkUs"))
{
printf("Checking our fa files for size\n");
checkOurDir(cgiString("checkUs"), contigList, hash);
}
else
{
problemCount += flagMd(mdFile, hash);
printf("Verifying sequence sizes in %s\n", faFile);
problemCount += flagFa(faFile, hash);
for (contig = contigList; contig != NULL; contig = contig->next)
{
if (contig->gotFa && contig->gotMd)
continue;
++problemCount;
if (!contig->gotFa)
printf("No %s in %s\n", contig->name, faFile);
if (!contig->gotMd)
printf("No %s in %s\n", contig->name, mdFile);
}
printf("%d problems detected\n", problemCount);
}
}
void stopUpload(struct sqlConnection *conn)
/* Try and stop current upload. */
{
char *url = trimSpaces(cgiString("url"));
cgiMakeHiddenVar("url", url);
struct edwSubmit *sub = edwMostRecentSubmission(conn, url);
if (sub == NULL)
{
/* Submission hasn't happened yet - remove it from job queue. */
unsigned edwSubmitJobId = 0;
int posInQueue = edwSubmitPositionInQueue(conn, url, &edwSubmitJobId);
if (posInQueue >= 0)
{
char query[256];
sqlSafef(query, sizeof(query), "delete from edwSubmitJob where id=%u", edwSubmitJobId);
sqlUpdate(conn, query);
printf("Removed submission from %s from job queue\n", url);
}
}
else
{
char query[256];
sqlSafef(query, sizeof(query),
"update edwSubmit set errorMessage='Stopped by user.' where id=%u", sub->id);
sqlUpdate(conn, query);
}
monitorSubmission(conn);
}
void doMiddle()
{
char *preIntron, *startIntron, *endIntron, *postIntron;
int count = 0;
int matchCount = 0;
int maxCount;
struct nameOff *matchList;
boolean invert = cgiVarExists("Invert");
preIntron = cgiString("preIntron");
startIntron = cgiString("startIntron");
endIntron = cgiString("endIntron");
postIntron = cgiString("postIntron");
/* Just for debugging cut search short if matches special magic */
maxCount = atoi(preIntron);
if (maxCount <= 0)
maxCount = 0x7ffffff;
eraseWhiteSpace(preIntron);
eraseWhiteSpace(startIntron);
eraseWhiteSpace(endIntron);
eraseWhiteSpace(postIntron);
tolowers(preIntron);
tolowers(startIntron);
tolowers(endIntron);
tolowers(postIntron);
matchList = scanIntronFile(preIntron, startIntron, endIntron, postIntron, invert);
if (matchList == NULL)
errAbort("Sorry, no matches to %s%s %s %s %s", (invert ? "inverted " : ""), preIntron, startIntron, endIntron, postIntron);
printf("<P>%d introns matching %s%s(%s %s)%s. ", slCount(matchList),
(invert ? "inverted " : ""), preIntron, startIntron, endIntron, postIntron);
printf("Shortcut to <A HREF=\"#1\">frequency counts.</A></P>");
htmlHorizontalLine();
showMatches(matchList);
htmlHorizontalLine();
printf("<TT><PRE>");
printf("<A NAME=\"1\"></A>");
printAllHistograms();
printf("</TT></PRE>");
}
struct edwScriptRegistry *edwScriptRegistryFromCgi()
/* Get script registery from cgi variables. Does authentication too. */
{
struct sqlConnection *conn = edwConnect();
char *user = sqlEscapeString(cgiString("user"));
char *password = sqlEscapeString(cgiString("password"));
char query[256];
sqlSafef(query, sizeof(query), "select * from edwScriptRegistry where name='%s'", user);
struct edwScriptRegistry *reg = edwScriptRegistryLoadByQuery(conn, query);
if (reg == NULL)
accessDenied();
char key[EDW_SID_SIZE];
edwMakeSid(password, key);
if (!sameString(reg->secretHash, key))
accessDenied();
sqlDisconnect(&conn);
return reg;
}
void doMiddle()
{
char *commandLine = cgiString("commandLine");
char *words[50];
int wordCount;
wordCount = chopString(commandLine, whiteSpaceChopper, words, ArraySize(words));
printf("<PRE>");
test(wordCount, words);
printf("</PRE>");
}
int main(int argc, char *argv[])
{
char *geneName;
char title[256];
if (argc == 2 && sameWord(argv[1], "test"))
putenv("QUERY_STRING=geneName=I:4000-5500&hiliteNear=0.917112&intronsLowerCase=On");
geneName = cgiString("geneName");
sprintf(title, "%s DNA Sequence", geneName);
dnaUtilOpen();
htmShell(title, doMiddle, "QUERY");
return 0;
}
void doMiddle()
/* Print middle parts of web page. Get database and transcript
* ID from CGI, and print info about that transcript. */
{
char *transId = cgiString("transId");
char *db = cgiString("db");
struct knownInfo *ki, *kiList = NULL;
struct sqlConnection *conn = sqlConnect(db);
struct sqlResult *sr;
char **row;
char query[256];
/* Get a list of all that have that ID. */
sqlSafef(query, sizeof query, "select * from knownInfo where transId = '%s'", transId);
sr = sqlGetResult(conn, query);
while ((row = sqlNextRow(sr)) != NULL)
{
ki = knownInfoLoad(row);
slAddHead(&kiList, ki);
}
sqlFreeResult(&sr);
slReverse(&kiList);
/* Print title that says how many match. */
printf("<H2>Transcript %s - %d match</H2>\n", transId, slCount(kiList));
/* Print some info for each match */
for (ki = kiList; ki != NULL; ki = ki->next)
{
printf("<B>geneId</B>: %s<BR>\n", ki->geneId);
printf("<B>geneName</B>: %s<BR>\n",
lookupName(conn, "geneName", ki->geneName));
/* ~~~ Todo: fill in other info. ~~~ */
}
knownInfoFreeList(&kiList);
sqlDisconnect(&conn);
}
void doMiddle()
/* Write HTML file to stdout. */
{
char *mapName;
if ((mapName = cgiOptionalString("map")) != NULL)
{
printf("<TT>\n");
doMap(mapName);
}
else if (cgiVarExists("contig"))
{
doFuzzyFind();
}
else
{
char *bacAcc = cgiString("bacAcc");
char *repeatMask = cgiString("repeatMask");
int trim = cgiInt("trim");
if (trim < 0)
trim = 0;
bacTrack(bacAcc, trim, repeatMask);
}
}
int main(int argc, char *argv[])
/* Process command line. */
{
struct hash *liftHash = NULL;
cgiSpoof(&argc, argv);
if (argc != 6)
usage();
if (cgiVarExists("spacing"))
chromInsertsSetDefaultGapSize(cgiInt("spacing"));
if (cgiVarExists("lift"))
liftHash = readLift(cgiString("lift"));
ctgToChromFa(argv[1], argv[2], argv[3], argv[4], argv[5], liftHash);
return 0;
}
int main(int argc, char *argv[])
{
struct altGraphX *agList = NULL;
int cassetteCount = 0;
float minConfidence = 0;
char *bedFileName = NULL;
char *faFile = NULL;
FILE *faOut = NULL;
FILE *bedOut = NULL;
boolean mrnaFilter = FALSE;
float estPrior = 0.0;
int minSize = 0;
if(argc < 4)
usage();
cgiSpoof(&argc, argv);
warn("Loading graphs.");
agList = altGraphXLoadAll(argv[1]);
bedFileName = cgiOptionalString("bedFile");
minConfidence = cgiDouble("minConf");
db = cgiString("db");
faFile = cgiOptionalString("faFile");
estPrior = cgiOptionalDouble("estPrior", 10);
minSize = cgiOptionalInt("minSize", 0);
mrnaFilter = cgiBoolean("mrnaFilter");
if(mrnaFilter)
loadMrnaHash();
warn("Counting cassette exons from %d clusters above confidence: %f", slCount(agList), minConfidence);
if(bedFileName != NULL)
{
bedOut = mustOpen(bedFileName, "w");
printCommandState(argc, argv, bedOut);
fprintf(bedOut, "track name=cass_conf-%4.2f_est-%3.2f description=\"spliceStats minConf=%4.2f estPrior=%3.2f minSize=%d\"\n",
minConfidence, estPrior, minConfidence, estPrior, minSize);
}
if(faFile != NULL)
faOut = mustOpen(faFile, "w");
cassetteCount = countCassetteExons(agList, minConfidence, faOut,bedOut );
carefulClose(&faOut);
carefulClose(&bedOut);
warn("%d cassette exons out of %d clusters in %s", cassetteCount, slCount(agList), argv[1]);
altGraphXFreeList(&agList);
return 0;
}
void submitUrl(struct sqlConnection *conn)
/* Submit validated manifest if it is not already in process. Show
* progress once it is in progress. */
{
/* Parse email and URL out of CGI vars. Do a tiny bit of error checking. */
char *url = trimSpaces(cgiString("url"));
if (!stringIn("://", url))
errAbort("%s doesn't seem to be a valid URL, no '://'", url);
/* Do some reality checks that email and URL actually exist. */
edwMustGetUserFromEmail(conn, userEmail);
int sd = netUrlMustOpenPastHeader(url);
close(sd);
edwAddSubmitJob(conn, userEmail, url, cgiBoolean("update"));
/* Give the system a half second to react and then put up status info about submission */
sleep1000(1000);
monitorSubmission(conn);
}
void autoXml(char *dtdxFile, char *outRoot)
/* autoXml - Generate structures code and parser for XML file from DTD-like spec. */
{
struct dtdElement *elList = NULL;
struct hash *elHash = NULL;
char hName[512], cName[512];
char outDir[256];
splitPath(outRoot, outDir, prefix, NULL);
if (cgiVarExists("prefix"))
strcpy(prefix, cgiString("prefix"));
if (outDir[0] != 0)
makeDir(outDir);
dtdParse(dtdxFile, prefix, textField, &elList, &elHash);
printf("Parsed %d elements in %s\n", slCount(elList), dtdxFile);
sprintf(hName, "%s.h", outRoot);
makeH(elList, hName);
sprintf(cName, "%s.c", outRoot);
makeC(elList, cName, hName);
printf("Generated code in %s\n", cName);
}
int main(int argc, char *argv[])
{
long enteredMainTime = clock1000();
measureTime(NULL);
browserName = hBrowserName();
organization = "UCSC";
/* Push very early error handling - this is just
* for the benefit of the cgiVarExists, which
* somehow can't be moved effectively into doMiddle. */
htmlPushEarlyHandlers();
cgiSpoof(&argc, argv);
char * link = webTimeStampedLinkToResourceOnFirstCall("HGStyle.css",TRUE); // resource file link
if (link) // wrapped in html
htmlSetStyle(link);
oldVars = hashNew(10);
if (cgiVarExists("hgt.redirectTool"))
{
printf("Content-type: text/html\n\n");
errAbortSetDoContentType(FALSE);
cart = cartForSession(hUserCookie(), NULL, NULL);
extToolRedirect(cart, cgiString("hgt.redirectTool"));
}
else
cartHtmlShell("UCSC Genome Browser v"CGI_VERSION, doMiddle, hUserCookie(), excludeVars, oldVars);
if (measureTiming)
measureTime("Time to write and close cart");
if (measureTiming)
{
fprintf(stdout, "<span class='timing'>Overall total time: %ld millis<br /></span>\n",
clock1000() - enteredMainTime);
}
cgiExitTime("hgTracks", enteredMainTime);
return 0;
}
int main(int argc, char *argv[])
/* Process command line. */
{
pushCarefulMemHandler(100000000);
cgiSpoof(&argc, argv);
htmlSetStyle(htmlStyleUndecoratedLink);
htmlSetBgColor(HG_CL_OUTSIDE);
oldCart = hashNew(10);
cart = cartAndCookie(hUserCookie(), excludeVars, oldCart);
getDbAndGenome(cart, &database, &genome, oldCart);
//hSetDb(database);
conn = hAllocConn(database);
database = strdup("h1n1");
/* Get sortOn. Revert to default by subject Id. */
orderOn = cartUsualString(cart, orderVarName, "+subjId");
displayCountString = cartUsualString(cart, countVarName, "50");
if (sameString(displayCountString, "all"))
displayCount = BIGNUM;
else
displayCount = atoi(displayCountString);
colList = getColumns(conn);
if (cgiVarExists("submit_filter"))
{
struct dyString *head = dyStringNew(1024);
boolean redir = cgiVarExists(redirectName);
struct subjInfo *subjList = NULL;
struct column *ordList = colList;
struct column *ord = curOrder(ordList);
subjList = getOrderedList(ord, colList, conn, BIGNUM);
saveSubjList(subjList);
if ((!subjList || redir))
{
if (subjList && redir)
{
dyStringPrintf(head,
"<META HTTP-EQUIV=\"REFRESH\" CONTENT=\"0;URL=/cgi-bin/%s\">"
"<META HTTP-EQUIV=\"Pragma\" CONTENT=\"no-cache\">"
"<META HTTP-EQUIV=\"Expires\" CONTENT=\"-1\">"
, cgiString(redirectName));
cartRemove(cart, redirectName);
}
htmStartWithHead(stdout, head->string, "GISAID Table View");
if (!subjList) /* if everything has been filtered out, we'll have to go back */
{
hPrintf("No subject(s) found with the filtering conditions specified.<br>");
hPrintf("Click <a href=\"gisaidTable?gisaidTable.do.advFilter=filter+%%28now+on%%29\">here</a> "
"to return to Select Subjects.<br>");
}
cartCheckout(&cart);
htmlEnd();
hFreeConn(&conn);
return 0;
}
}
htmStart(stdout, "GISAID Table View");
cartWarnCatcher(doMiddle, cart, htmlVaWarn);
cartCheckout(&cart);
htmlEnd();
hFreeConn(&conn);
return 0;
}
int main(int argc, char *argv[])
{
long enteredMainTime = clock1000();
struct dyString *output = newDyString(10000);
setUdcCacheDir();
cgiSpoof(&argc, argv);
pushWarnHandler(htmlVaBadRequestAbort);
pushAbortHandler(htmlVaBadRequestAbort);
char *database = cgiString("db");
char *cmd = cgiString("cmd");
char *jsonp = cgiOptionalString("jsonp");
if (!hDbExists(database))
errAbort("Invalid database '%s'", database);
if (!strcmp(cmd, "defaultPos"))
{
dyStringPrintf(output, "{\"pos\": \"%s\"}", hDefaultPos(database));
}
else if (!strcmp(cmd, "metaDb"))
{
// Return list of values for given metaDb var
// e.g. http://genome.ucsc.edu/hgApi?db=hg18&cmd=metaDb&var=cell
struct sqlConnection *conn = hAllocConn(database);
boolean metaDbExists = sqlTableExists(conn, "metaDb");
if (metaDbExists)
{
char *var = cgiOptionalString("var");
if (!var)
errAbort("Missing var parameter");
boolean fileSearch = (cgiOptionalInt("fileSearch",0) == 1);
struct slPair *pairs = mdbValLabelSearch(conn, var, MDB_VAL_STD_TRUNCATION, FALSE,
!fileSearch, fileSearch);
struct slPair *pair;
dyStringPrintf(output, "[\n");
for (pair = pairs; pair != NULL; pair = pair->next)
{
if (pair != pairs)
dyStringPrintf(output, ",\n");
dyStringPrintf(output, "['%s','%s']", javaScriptLiteralEncode(mdbPairLabel(pair)),
javaScriptLiteralEncode(mdbPairVal(pair)));
}
dyStringPrintf(output, "\n]\n");
}
else
errAbort("Assembly does not support metaDb");
}
// TODO: move to lib since hgTracks and hgApi share
#define METADATA_VALUE_PREFIX "hgt_mdbVal"
else if (startsWith(METADATA_VALUE_PREFIX, cmd))
{
// Returns metaDb value control: drop down or free text, with or without help link.
// e.g. http://genome.ucsc.edu/hgApi?db=hg18&cmd=hgt_mdbVal3&var=cell
// TODO: Move guts to lib, so that hgTracks::searchTracks.c and hgApi.c can share
struct sqlConnection *conn = hAllocConn(database);
boolean metaDbExists = sqlTableExists(conn, "metaDb");
if (metaDbExists)
{
char *var = cgiOptionalString("var");
if (!var)
errAbort("Missing var parameter");
int ix = atoi(cmd+strlen(METADATA_VALUE_PREFIX)); // 1 based index
if (ix == 0) //
errAbort("Unsupported 'cmd' parameter");
enum cvSearchable searchBy = cvSearchMethod(var);
char name[128];
safef(name,sizeof name,"%s%i",METADATA_VALUE_PREFIX,ix);
if (searchBy == cvSearchBySingleSelect || searchBy == cvSearchByMultiSelect)
{
boolean fileSearch = (cgiOptionalInt("fileSearch",0) == 1);
struct slPair *pairs = mdbValLabelSearch(conn, var, MDB_VAL_STD_TRUNCATION, FALSE,
!fileSearch, fileSearch);
if (slCount(pairs) > 0)
{
char *dropDownHtml =
cgiMakeSelectDropList((searchBy == cvSearchByMultiSelect),
name, pairs, NULL, ANYLABEL, "mdbVal",
"style='min-width: 200px; font-size: .9em;' "
"onchange='findTracksMdbValChanged(this);'");
if (dropDownHtml)
{
dyStringAppend(output,dropDownHtml);
freeMem(dropDownHtml);
}
slPairFreeList(&pairs);
}
}
else if (searchBy == cvSearchByFreeText)
{
dyStringPrintf(output,"<input type='text' name='%s' value='' class='mdbVal freeText' "
"onchange='findTracksMdbValChanged(this);' style='max-width:310px; "
"width:310px; font-size:.9em;'>", name);
}
else if (searchBy == cvSearchByWildList)
{
dyStringPrintf(output,"<input type='text' name='%s' value='' class='mdbVal wildList' "
"title='enter comma separated list of values' "
"onchange='findTracksMdbValChanged(this);' style='max-width:310px; "
"width:310px; font-size:.9em;'>", name);
}
else if (searchBy == cvSearchByDateRange || searchBy == cvSearchByIntegerRange)
{
// TO BE IMPLEMENTED
}
else
errAbort("Metadata variable not searchable");
dyStringPrintf(output,"<span id='helpLink%i'> </span>",ix);
}
else
errAbort("Assembly does not support metaDb");
}
else if (!strcmp(cmd, "tableMetadata"))
{ // returns an html table with metadata for a given track
char *trackName = cgiOptionalString("track");
boolean showLonglabel = (NULL != cgiOptionalString("showLonglabel"));
boolean showShortLabel = (NULL != cgiOptionalString("showShortLabel"));
if (trackName != NULL)
{
// hTrackDbForTrackAndAncestors avoids overhead of getting whole track list!
struct trackDb *tdb = hTrackDbForTrackAndAncestors(database, trackName);
if (tdb != NULL)
{
char * html = metadataAsHtmlTable(database,tdb,showLonglabel,showShortLabel);
if (html)
{
dyStringAppend(output,html);
freeMem(html);
}
else
dyStringPrintf(output,"No metadata found for track %s.",trackName);
}
else
dyStringPrintf(output,"Track %s not found",trackName);
}
else
dyStringAppend(output,"No track variable found");
}
else if (sameString(cmd, "codonToPos") || sameString(cmd, "exonToPos"))
{
char query[256];
struct sqlResult *sr;
char **row;
struct genePred *gp;
char *name = cgiString("name");
char *table = cgiString("table");
int num = cgiInt("num");
struct sqlConnection *conn = hAllocConn(database);
sqlSafef(query, sizeof(query), "select name, chrom, strand, txStart, txEnd, cdsStart, cdsEnd, exonCount, exonStarts, exonEnds from %s where name = '%s'", table, name);
sr = sqlGetResult(conn, query);
if ((row = sqlNextRow(sr)) != NULL)
{
gp = genePredLoad(row);
boolean found;
int start, end;
if (sameString(cmd, "codonToPos"))
found = codonToPos(gp, num, &start, &end);
else
found = exonToPos(gp, num, &start, &end);
if (found)
dyStringPrintf(output, "{\"pos\": \"%s:%d-%d\"}", gp->chrom, start + 1, end);
else
dyStringPrintf(output, "{\"error\": \"%d is an invalid %s for this gene\"}", num, sameString(cmd, "codonToPos") ? "codon" : "exon");
}
else
dyStringPrintf(output, "{\"error\": \"Couldn't find item: %s\"}", name);
sqlFreeResult(&sr);
hFreeConn(&conn);
}
else
{
warn("unknown cmd: %s",cmd);
errAbort("Unsupported 'cmd' parameter");
}
apiOut(dyStringContents(output), jsonp);
cgiExitTime("hgApi", enteredMainTime);
return 0;
}
void doMiddle()
{
char *seqName;
boolean intronsLowerCase = TRUE;
boolean intronsParenthesized = FALSE;
boolean hiliteNear = FALSE;
int startRange = 0;
int endRange = 0;
boolean gotRange = FALSE;
struct dnaSeq *cdnaSeq;
boolean isChromRange = FALSE;
DNA *dna;
char *translation = NULL;
seqName = cgiString("geneName");
seqName = trimSpaces(seqName);
if (cgiVarExists("intronsLowerCase"))
intronsLowerCase = cgiBoolean("intronsLowerCase");
if (cgiVarExists("intronsParenthesized"))
intronsParenthesized = cgiBoolean("intronsParenthesized");
if (cgiVarExists("startRange") && cgiVarExists("endRange" ))
{
startRange = cgiInt("startRange");
endRange = cgiInt("endRange");
gotRange = TRUE;
}
if (cgiVarExists("hiliteNear"))
{
hiliteNear = TRUE;
}
fprintf(stdout, "<P><TT>\n");
/* The logic here is a little complex to optimize speed.
* If we can decide what type of thing the name refers to by
* simply looking at the name we do. Otherwise we have to
* search the database in various ways until we get a hit. */
if (wormIsNamelessCluster(seqName))
{
isChromRange = TRUE;
}
else if (wormIsChromRange(seqName))
{
isChromRange = TRUE;
}
else if (getWormGeneDna(seqName, &dna, TRUE))
{
if (cgiBoolean("litLink"))
{
char nameBuf[64];
char *geneName = NULL;
char *productName = NULL;
char *coding;
int transSize;
struct wormCdnaInfo info;
printf("<H3>Information and Links for %s</H3>\n", seqName);
if (wormInfoForGene(seqName, &info))
{
if (info.description)
printf("<P>%s</P>\n", info.description);
geneName = info.gene;
productName = info.product;
}
else
{
if (wormIsGeneName(seqName))
geneName = seqName;
else if (wormGeneForOrf(seqName, nameBuf, sizeof(nameBuf)))
geneName = nameBuf;
}
coding = cloneUpperOnly(dna);
transSize = 1 + (strlen(coding)+2)/3;
translation = needMem(1+strlen(coding)/3);
dnaTranslateSome(coding, translation, transSize);
freez(&coding);
if (geneName)
{
printf("<A HREF=\"http://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?form=4&db=m"
"&term=C+elegans+%s&dispmax=50&relentrezdate=No+Limit\">", geneName);
printf("PubMed search on gene: </A>%s<BR>\n", geneName);
}
if (productName)
{
char *encoded = cgiEncode(productName);
printf("<A HREF=\"http://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?form=4&db=m"
"&term=%s&dispmax=50&relentrezdate=No+Limit\">", encoded);
printf("PubMed search on product:</A> %s<BR>\n", productName);
freeMem(encoded);
}
/* Process name to get rid of isoform letter for Proteome. */
if (geneName)
strcpy(nameBuf, geneName);
else
{
strcpy(nameBuf, seqName);
#ifdef NEVER
/* Sometimes Proteome requires the letter after the orf name
* in alt-spliced cases, sometimes it can't handle it.... */
nameLen = strlen(nameBuf);
if (wormIsOrfName(nameBuf) && isalpha(nameBuf[nameLen-1]))
{
char *dotPos = strrchr(nameBuf, '.');
if (dotPos != NULL && isdigit(dotPos[1]))
nameBuf[nameLen-1] = 0;
}
#endif /* NEVER */
}
printf("<A HREF=\"http://www.wormbase.org/db/seq/sequence?name=%s;class=Sequence\">", seqName);
printf("WormBase link on:</A> %s<BR>\n", seqName);
printf("<A HREF=\"http://www.proteome.com/databases/WormPD/reports/%s.html\">", nameBuf);
printf("Proteome link on:</A> %s<BR>\n<BR>\n", nameBuf);
printf("<A HREF=#DNA>Genomic DNA Sequence</A><BR>\n");
if (hiliteNear)
printf("<A HREF=\"#CLICKED\">Shortcut to where you clicked in gene</A><BR>");
printf("<A HREF=#protein>Translated Protein Sequence</A><BR>\n");
htmlHorizontalLine();
printf("<A NAME=DNA></A>");
printf("<H3>%s Genomic DNA sequence</H3>", seqName);
}
if (!intronsLowerCase)
tolowers(dna);
if (hiliteNear)
{
if (!gotRange)
{
double nearPos = cgiDouble("hiliteNear");
int rad = 5;
int dnaSize = strlen(dna);
long mid = (int)(dnaSize * nearPos);
startRange = mid - rad;
if (startRange < 0) startRange = 0;
endRange = mid + rad;
if (endRange >= dnaSize) endRange = dnaSize - 1;
}
}
outputSeq(dna, strlen(dna), hiliteNear, startRange, endRange, stdout);
freez(&dna);
}
else if (wormCdnaSeq(seqName, &cdnaSeq, NULL))
{
outputSeq(cdnaSeq->dna, cdnaSeq->size, FALSE, 0, 0, stdout);
}
else
{
isChromRange = TRUE;
}
if (isChromRange)
{
char *chromId;
int start, end;
char strand = '+';
int size;
if (!wormGeneRange(seqName, &chromId, &strand, &start, &end))
errAbort("Can't find %s",seqName);
size = end - start;
if (intronsLowerCase)
dna = wormChromPartExonsUpper(chromId, start, size);
else
{
dna = wormChromPart(chromId, start, size);
touppers(dna);
}
if (cgiVarExists("strand"))
strand = cgiString("strand")[0];
if (strand == '-')
reverseComplement(dna, size);
outputSeq(dna, size, FALSE, 0, 0, stdout);
}
if (translation != NULL)
{
htmlHorizontalLine();
printf("<A NAME=protein></A>");
printf("<H3>Translated Protein of %s</H3>\n", seqName);
outputSeq(translation, strlen(translation), FALSE, 0, 0, stdout);
freez(&translation);
}
fprintf(stdout, "</TT></P>\n");
}
void blatSeq(char *userSeq, char *organism)
/* Blat sequence user pasted in. */
{
FILE *f;
struct dnaSeq *seqList = NULL, *seq;
struct tempName pslTn, faTn;
int maxSingleSize, maxTotalSize, maxSeqCount;
int minSingleSize = minMatchShown;
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;
boolean txTxBoth = FALSE;
struct gfOutput *gvo;
boolean qIsProt = FALSE;
enum gfType qType, tType;
struct hash *tFileCache = gfFileCacheNew();
boolean feelingLucky = cgiBoolean("Lucky");
getDbAndGenome(cart, &db, &genome, oldVars);
if(!feelingLucky)
cartWebStart(cart, db, "%s BLAT Results", trackHubSkipHubName(organism));
/* Load user sequence and figure out if it is DNA or protein. */
if (sameWord(type, "DNA"))
{
seqList = faSeqListFromMemText(seqLetters, TRUE);
uToT(seqList);
isTx = FALSE;
}
else if (sameWord(type, "translated RNA") || sameWord(type, "translated DNA"))
{
seqList = faSeqListFromMemText(seqLetters, TRUE);
uToT(seqList);
isTx = TRUE;
isTxTx = TRUE;
txTxBoth = sameWord(type, "translated DNA");
}
else if (sameWord(type, "protein"))
{
seqList = faSeqListFromMemText(seqLetters, FALSE);
isTx = TRUE;
qIsProt = TRUE;
}
else
{
seqList = faSeqListFromMemTextRaw(seqLetters);
isTx = !seqIsDna(seqList);
if (!isTx)
{
for (seq = seqList; seq != NULL; seq = seq->next)
{
seq->size = dnaFilteredSize(seq->dna);
dnaFilter(seq->dna, seq->dna);
toLowerN(seq->dna, seq->size);
subChar(seq->dna, 'u', 't');
}
}
else
{
for (seq = seqList; seq != NULL; seq = seq->next)
{
seq->size = aaFilteredSize(seq->dna);
aaFilter(seq->dna, seq->dna);
toUpperN(seq->dna, seq->size);
}
qIsProt = TRUE;
}
}
if (seqList != NULL && seqList->name[0] == 0)
{
freeMem(seqList->name);
seqList->name = cloneString("YourSeq");
}
trimUniq(seqList);
/* If feeling lucky only do the first on. */
if(feelingLucky && seqList != NULL)
{
seqList->next = NULL;
}
/* Figure out size allowed. */
maxSingleSize = (isTx ? 10000 : 75000);
maxTotalSize = maxSingleSize * 2.5;
#ifdef LOWELAB
maxSeqCount = 200;
#else
maxSeqCount = 25;
#endif
/* Create temporary file to store sequence. */
trashDirFile(&faTn, "hgSs", "hgSs", ".fa");
faWriteAll(faTn.forCgi, seqList);
/* Create a temporary .psl file with the alignments against genome. */
trashDirFile(&pslTn, "hgSs", "hgSs", ".pslx");
f = mustOpen(pslTn.forCgi, "w");
gvo = gfOutputPsl(0, qIsProt, FALSE, f, FALSE, TRUE);
serve = findServer(db, isTx);
/* Write header for extended (possibly protein) psl file. */
if (isTx)
{
if (isTxTx)
{
qType = gftDnaX;
tType = gftDnaX;
}
else
{
qType = gftProt;
tType = gftDnaX;
}
}
else
{
qType = gftDna;
tType = gftDna;
}
pslxWriteHead(f, qType, tType);
if (qType == gftProt)
{
minSingleSize = 14;
}
else if (qType == gftDnaX)
{
minSingleSize = 36;
}
/* 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);
if ((seqCount&1) == 0) // Call bot delay every 2nd time starting with first time
hgBotDelay();
if (++seqCount > maxSeqCount)
{
warn("More than 25 input sequences, stopping at %s.",
seq->name);
break;
}
if (oneSize > maxSingleSize)
{
warn("Sequence %s is %d letters long (max is %d), skipping",
seq->name, oneSize, maxSingleSize);
continue;
}
if (oneSize < minSingleSize)
{
warn("Warning: Sequence %s is only %d letters long (%d is the recommended minimum)",
seq->name, oneSize, minSingleSize);
// we could use "continue;" here to actually enforce skipping,
// but let's give the short sequence a chance, it might work.
// minimum possible length = tileSize+stepSize, so mpl=16 for dna stepSize=5, mpl=10 for protein.
if (qIsProt && oneSize < 1) // protein does not tolerate oneSize==0
continue;
}
totalSize += oneSize;
if (totalSize > maxTotalSize)
{
warn("Sequence %s would take us over the %d letter limit, stopping here.",
seq->name, maxTotalSize);
break;
}
conn = gfConnect(serve->host, serve->port);
if (isTx)
{
gvo->reportTargetStrand = TRUE;
if (isTxTx)
{
gfAlignTransTrans(&conn, serve->nibDir, seq, FALSE, 5,
tFileCache, gvo, !txTxBoth);
if (txTxBoth)
{
reverseComplement(seq->dna, seq->size);
conn = gfConnect(serve->host, serve->port);
gfAlignTransTrans(&conn, serve->nibDir, seq, TRUE, 5,
tFileCache, gvo, FALSE);
}
}
else
{
gfAlignTrans(&conn, serve->nibDir, seq, 5, tFileCache, gvo);
}
}
else
{
gfAlignStrand(&conn, serve->nibDir, seq, FALSE, minMatchShown, tFileCache, gvo);
reverseComplement(seq->dna, seq->size);
conn = gfConnect(serve->host, serve->port);
gfAlignStrand(&conn, serve->nibDir, seq, TRUE, minMatchShown, tFileCache, gvo);
}
gfOutputQuery(gvo, f);
}
carefulClose(&f);
showAliPlaces(pslTn.forCgi, faTn.forCgi, serve->db, qType, tType,
organism, feelingLucky);
if(!feelingLucky)
cartWebEnd();
gfFileCacheFree(&tFileCache);
}
void doMiddle()
/* Write what goes between BODY and /BODY */
{
pushWarnHandler(localWarn);
if (!cgiServerHttpsIsOn())
usage();
struct sqlConnection *conn = edwConnectReadWrite();
printf("<FORM ACTION=\"edwWebRegisterScript\" METHOD=POST>\n");
printf("<B>Register Script with ENCODE Data Warehouse</B><BR>\n");
#ifdef SOON
uglyf("HTTP_AUTHENTICATION: '%s'<BR>\n", getenv("HTTP_AUTHENTICATION"));
uglyf("HTTP_AUTHORIZATION: '%s'<BR>\n", getenv("HTTP_AUTHORIZATION"));
dumpEnv(mainEnv);
#endif
if (userEmail == NULL)
{
printf("Please sign in:");
printf("<INPUT TYPE=BUTTON NAME=\"signIn\" VALUE=\"sign in\" id=\"signin\">");
}
else if (cgiVarExists("description"))
{
struct edwUser *user = edwUserFromEmail(conn, userEmail);
if (user == NULL)
edwWarnUnregisteredUser(userEmail);
else
{
char password[HEXED_32_SIZE];
edwRandomHexed32(password);
char babyName[HEXED_32_SIZE];
edwRandomBabble(babyName, sizeof(babyName));
edwRegisterScript(conn, user, babyName, password, cgiString("description"));
printf("Script now registered.<BR>\n");
printf("The script user name is %s.<BR>\n", babyName);
printf("The script password is %s.<BR>\n", password);
printf("Please save the script user name and password somewhere. ");
puts("Please pass these two and the URL");
puts(" of your validated manifest file (validated.txt) to our server to submit data.");
puts("Construct a URL of the form:<BR>");
printf("<PRE>https://encodedcc.sdsc.edu/cgi-bin/edwScriptSubmit"
"?user=%s&password=%s&url=%s\n</PRE>",
babyName, password,
cgiEncode("http://your.host.edu/your_dir/validated.txt"));
puts("That is pass the CGI encoded variables user, password, and url to the ");
puts("web services CGI at");
puts("https://encodedcc.sdsc.edu/cgi-bin/edwScriptSubmit. ");
puts("You can use the http://encodedcc.sdsc.edu/cgi-bin/edwWebBrowse site to ");
puts("monitor your submission interactively. Please contact your wrangler if you ");
puts("have any questions.<BR>");
cgiMakeButton("submit", "Register another script");
}
printf(" ");
edwPrintLogOutButton();
}
else
{
struct edwUser *user = edwUserFromEmail(conn, userEmail);
edwPrintLogOutButton();
if (user == NULL)
edwWarnUnregisteredUser(userEmail);
else
{
printf("%s is authorized to register a new script<BR>\n", userEmail);
printf("<BR>Script description:\n");
cgiMakeTextVar("description", NULL, 80);
cgiMakeSubmitButton();
}
}
printf("</FORM>\n");
}
void edwScriptSubmitStatus()
/* edwScriptSubmitStatus - Programatically check status of submission.. */
{
/* Pause a second - prevent inadvertent harsh denial of service from scripts. */
sleep(2);
edwScriptRegistryFromCgi();
/* Get submission from url. */
struct sqlConnection *conn = edwConnect();
char query[512];
char *url = cgiString("url");
struct edwSubmit *sub = edwMostRecentSubmission(conn, url);
char *status = NULL;
if (sub == NULL)
{
int posInQueue = edwSubmitPositionInQueue(conn, url, NULL);
if (posInQueue == -1)
errAbort("%s has not been submitted", url);
else
status = "pending";
}
else
{
time_t endUploadTime = sub->endUploadTime;
if (!isEmpty(sub->errorMessage))
{
status = "error";
}
else if (endUploadTime == 0)
{
status = "uploading";
}
else
{
safef(query, sizeof(query),
"select count(*) from edwFile where submitId=%u and errorMessage != ''",
sub->id);
int errCount = sqlQuickNum(conn, query);
int newValid = edwSubmitCountNewValid(sub, conn);
if (newValid + errCount < sub->newFiles)
status = "validating";
else if (errCount > 0)
status = "error";
else
status = "success";
}
}
/* Construct JSON result */
struct dyString *dy = dyStringNew(0);
dyStringPrintf(dy, "{\n");
dyStringPrintf(dy, " \"status\": \"%s\"", status);
if (sameString(status, "error"))
{
dyStringPrintf(dy, ",\n");
dyStringPrintf(dy, " \"errors\": [\n");
int errCount = 0;
if (!isEmpty(sub->errorMessage))
{
addErrFile(dy, errCount, sub->url, sub->errorMessage);
++errCount;
}
safef(query, sizeof(query), "select * from edwFile where submitId=%u and errorMessage != ''",
sub->id);
struct edwFile *file, *fileList = edwFileLoadByQuery(conn, query);
for (file = fileList; file != NULL; file = file->next)
{
addErrFile(dy, errCount, file->submitFileName, file->errorMessage);
++errCount;
}
dyStringPrintf(dy, "\n ]\n");
dyStringPrintf(dy, "}\n");
}
else
{
dyStringPrintf(dy, "\n}\n");
}
/* Write out HTTP response */
printf("Content-Length: %d\r\n", dy->stringSize);
puts("Content-Type: application/json; charset=UTF-8\r");
puts("\r");
printf("%s", dy->string);
}
void doMiddle(struct cart *theCart)
/* Set up globals and make web page */
{
/* struct liftOverChain *chainList = NULL, *chain; */
char *userData;
/* char *dataFile; */
char *dataFormat;
char *organism;
char *db;
float minBlocks, minMatch;
boolean multiple, fudgeThick;
int minSizeQ, minSizeT;
boolean refreshOnly = FALSE;
/* char *err = NULL; */
struct liftOverChain *chainList = NULL, *choice;
cart = theCart;
if (cgiOptionalString(HGLFT_ERRORHELP_VAR))
{
puts("<PRE>");
puts(liftOverErrHelp());
//system("/usr/bin/cal");
puts("</PRE>");
return;
}
/* Get data to convert - from userData variable, or if
* that is empty from a file. */
if (cartOptionalString(cart, "SubmitFile"))
userData = cartOptionalString(cart, HGLFT_DATAFILE_VAR);
else
userData = cartOptionalString(cart, HGLFT_USERDATA_VAR);
dataFormat = cartCgiUsualString(cart, HGLFT_DATAFORMAT_VAR, DEFAULT_FORMAT);
cartWebStart(cart, NULL, "Lift Genome Annotations");
getDbAndGenome(cart, &db, &organism, oldVars);
chainList = liftOverChainListFiltered();
choice = defaultChoices(chainList, db);
if (choice == NULL)
errAbort("Sorry, no conversions available from this assembly\n");
minSizeQ = cartCgiUsualInt(cart, HGLFT_MINSIZEQ, choice->minSizeQ);
minSizeT = cartCgiUsualInt(cart, HGLFT_MINSIZET, choice->minSizeT);
minBlocks = cartCgiUsualDouble(cart, HGLFT_MINBLOCKS, choice->minBlocks);
minMatch = cartCgiUsualDouble(cart, HGLFT_MINMATCH, choice->minMatch);
fudgeThick = cartCgiUsualBoolean(cart, HGLFT_FUDGETHICK, (choice->fudgeThick[0]=='Y') ? TRUE : FALSE);
multiple = cartCgiUsualBoolean(cart, HGLFT_MULTIPLE, (choice->multiple[0]=='Y') ? TRUE : FALSE);
refreshOnly = cartCgiUsualInt(cart, HGLFT_REFRESHONLY_VAR, 0);
webMain(choice, dataFormat, multiple);
liftOverChainFreeList(&chainList);
if (!refreshOnly && userData != NULL && userData[0] != '\0')
{
struct hash *chainHash = newHash(0);
char *chainFile;
struct tempName oldTn, mappedTn, unmappedTn;
FILE *old, *mapped, *unmapped;
char *line;
int lineSize;
char *fromDb, *toDb;
int ct = 0, errCt = 0;
/* read in user data and save to file */
makeTempName(&oldTn, HGLFT, ".user");
old = mustOpen(oldTn.forCgi, "w");
fputs(userData, old);
fputs("\n", old); /* in case user doesn't end last line */
carefulClose(&old);
chmod(oldTn.forCgi, 0666);
/* setup output files -- one for converted lines, the other
* for lines that could not be mapped */
makeTempName(&mappedTn, HGLFT, ".bed");
makeTempName(&unmappedTn, HGLFT, ".err");
mapped = mustOpen(mappedTn.forCgi, "w");
chmod(mappedTn.forCgi, 0666);
unmapped = mustOpen(unmappedTn.forCgi, "w");
chmod(unmappedTn.forCgi, 0666);
fromDb = cgiString(HGLFT_FROMDB_VAR);
toDb = cgiString(HGLFT_TODB_VAR);
chainFile = liftOverChainFile(fromDb, toDb);
if (chainFile == NULL)
errAbort("ERROR: Can't convert from %s to %s: no chain file loaded",
fromDb, toDb);
readLiftOverMap(chainFile, chainHash);
if (sameString(dataFormat, WIGGLE_FORMAT))
/* TODO: implement Wiggle */
{}
else if (sameString(dataFormat, POSITION_FORMAT))
{
/* minSizeT here and in liftOverChain.c/h has been renamed minChainT in liftOver.c */
/* ignore multiple, it must be false when position is used */
ct = liftOverPositions(oldTn.forCgi, chainHash,
minMatch, minBlocks, 0, minSizeQ,
minSizeT, 0,
fudgeThick, mapped, unmapped, FALSE, NULL, &errCt);
}
else if (sameString(dataFormat, BED_FORMAT))
{
/* minSizeT here and in liftOverChain.c/h has been renamed minChainT in liftOver.c */
ct = liftOverBed(oldTn.forCgi, chainHash,
minMatch, minBlocks, 0, minSizeQ,
minSizeT, 0,
fudgeThick, mapped, unmapped, multiple, NULL, &errCt);
}
else
/* programming error */
errAbort("ERROR: Unsupported data format: %s\n", dataFormat);
webNewSection("Results");
if (ct)
{
/* some records succesfully converted */
cgiParagraph("");
printf("Successfully converted %d record", ct);
printf("%s: ", ct > 1 ? "s" : "");
printf("<A HREF=%s TARGET=_blank>View Conversions</A>\n", mappedTn.forCgi);
}
if (errCt)
{
/* some records not converted */
cgiParagraph("");
printf("Conversion failed on %d record", errCt);
printf("%s. ", errCt > 1 ? "s" : "");
printf("<A HREF=%s TARGET=_blank>Display failure file</A> \n",
unmappedTn.forCgi);
printf("<A HREF=\"../cgi-bin/hgLiftOver?%s=1\" TARGET=_blank>Explain failure messages</A>\n", HGLFT_ERRORHELP_VAR);
puts("<P>Failed input regions:\n");
struct lineFile *errFile = lineFileOpen(unmappedTn.forCgi, TRUE);
puts("<BLOCKQUOTE><PRE>\n");
while (lineFileNext(errFile, &line, &lineSize))
puts(line);
lineFileClose(&errFile);
puts("</PRE></BLOCKQUOTE>\n");
}
if (sameString(dataFormat, POSITION_FORMAT) && multiple)
{
puts("<BLOCKQUOTE><PRE>\n");
puts("Note: multiple checkbox ignored since it is not supported for position format.");
puts("</PRE></BLOCKQUOTE>\n");
}
carefulClose(&unmapped);
}
webDataFormats();
webDownloads();
cartWebEnd();
}
void tryToDeprecate(struct sqlConnection *conn)
/* CGI variables are set - if possible deprecate, otherwise put up error message. */
{
pushWarnHandler(localWarn);
fileList = cgiString("fileList");
reason = cloneString(trimSpaces(cgiString("reason")));
if (isEmpty(reason))
{
warn("Please enter a reason for deprecation.");
getFileListAndReason(conn);
}
else
{
/* Go through list of accessions and make sure they are all well formed and correspond to files that exist. */
boolean ok = TRUE;
struct slName *accList = slNameListOfUniqueWords(cloneString(fileList), FALSE);
struct slName *acc;
struct slInt *idList = NULL, *idEl;
for (acc = accList; acc != NULL; acc = acc->next)
{
char *licensePlate = acc->name;
if (!startsWith(edwLicensePlatePrefix, licensePlate))
{
ok = FALSE;
warn("%s is not an accession, doesn't start with %s", licensePlate, edwLicensePlatePrefix);
break;
}
char query[256];
sqlSafef(query, sizeof(query), "select fileId from edwValidFile where licensePlate='%s'", licensePlate);
int id = sqlQuickNum(conn, query);
if (id == 0)
{
ok = FALSE;
warn("%s - no such accession. ", licensePlate);
break;
}
/* check to see is it ok tor deprecate this file */
if (!okToDeprecateThisFile(conn, id, userEmail))
{
ok = FALSE;
warn("You can not deprecate %s which was originally uploaded by %s.\n",
licensePlate, edwFindOwnerNameFromFileId(conn, id));
warn("Please click the check box below to override this rule.");
break;
}
idEl = slIntNew(id);
slAddTail(&idList, idEl);
}
if (accList == NULL)
{
warn("Please enter some file accessions");
ok = FALSE;
}
/* If a problem then put up page to try again, otherwise do deprecation. */
if (!ok)
getFileListAndReason(conn);
else
{
deprecateFileList(conn, idList, reason);
printf("Deprecated %d files<BR>\n", slCount(idList));
cgiMakeButton("submit", "Deprecate More Files");
printf(" ");
edwPrintLogOutButton();
}
}
}
static void hgSeqConcatRegionsDb(char *db, char *chrom, int chromSize, char strand, char *name,
int rCount, unsigned *rStarts, unsigned *rSizes,
boolean *exonFlags, boolean *cdsFlags)
/* Concatenate and print out dna for a series of regions. */
{
// Note: this code use to generate different sequence ids if the global
// database in hdb was different than the db parameter. This functionality
// has been removed since the global database was removed and it didn't
// appear to be used.
struct dnaSeq *rSeq = NULL;
struct dnaSeq *cSeq = NULL;
char recName[256];
int seqStart, seqEnd;
int offset, cSize;
int i;
boolean isRc = (strand == '-') || cgiBoolean("hgSeq.revComp");
boolean maskRep = cgiBoolean("hgSeq.maskRepeats");
int padding5 = cgiOptionalInt("hgSeq.padding5", 0);
int padding3 = cgiOptionalInt("hgSeq.padding3", 0);
char *casing = cgiString("hgSeq.casing");
char *repMasking = cgiString("hgSeq.repMasking");
char *granularity = cgiOptionalString("hgSeq.granularity");
boolean concatRegions = granularity && sameString("gene", granularity);
if (rCount < 1)
return;
/* Don't support padding if granularity is gene (i.e. concat'ing all). */
if (concatRegions)
{
padding5 = padding3 = 0;
}
i = rCount - 1;
seqStart = rStarts[0] - (isRc ? padding3 : padding5);
seqEnd = rStarts[i] + rSizes[i] + (isRc ? padding5 : padding3);
/* Padding might push us off the edge of the chrom; if so, truncate: */
if (seqStart < 0)
{
if (isRc)
padding3 += seqStart;
else
padding5 += seqStart;
seqStart = 0;
}
/* if we know the chromSize, don't pad out beyond it */
if ((chromSize > 0) && (seqEnd > chromSize))
{
if (isRc)
padding5 += (chromSize - seqEnd);
else
padding3 += (chromSize - seqEnd);
seqEnd = chromSize;
}
if (seqEnd <= seqStart)
{
printf("# Null range for %s_%s (range=%s:%d-%d 5'pad=%d 3'pad=%d) (may indicate a query-side insert)\n",
db,
name,
chrom, seqStart+1, seqEnd,
padding5, padding3);
return;
}
if (maskRep)
{
rSeq = hDnaFromSeq(db, chrom, seqStart, seqEnd, dnaMixed);
if (sameString(repMasking, "N"))
lowerToN(rSeq->dna, strlen(rSeq->dna));
if (!sameString(casing, "upper"))
tolowers(rSeq->dna);
}
else if (sameString(casing, "upper"))
rSeq = hDnaFromSeq(db, chrom, seqStart, seqEnd, dnaUpper);
else
rSeq = hDnaFromSeq(db, chrom, seqStart, seqEnd, dnaLower);
/* Handle casing and compute size of concatenated sequence */
cSize = 0;
for (i=0; i < rCount; i++)
{
if ((sameString(casing, "exon") && exonFlags[i]) ||
(sameString(casing, "cds") && cdsFlags[i]))
{
int rStart = rStarts[i] - seqStart;
toUpperN(rSeq->dna+rStart, rSizes[i]);
}
cSize += rSizes[i];
}
cSize += (padding5 + padding3);
AllocVar(cSeq);
cSeq->dna = needLargeMem(cSize+1);
cSeq->size = cSize;
offset = 0;
for (i=0; i < rCount; i++)
{
int start = rStarts[i] - seqStart;
int size = rSizes[i];
if (i == 0)
{
start -= (isRc ? padding3 : padding5);
assert(start == 0);
size += (isRc ? padding3 : padding5);
}
if (i == rCount-1)
{
size += (isRc ? padding5 : padding3);
}
memcpy(cSeq->dna+offset, rSeq->dna+start, size);
offset += size;
}
assert(offset == cSeq->size);
cSeq->dna[offset] = 0;
freeDnaSeq(&rSeq);
if (isRc)
reverseComplement(cSeq->dna, cSeq->size);
safef(recName, sizeof(recName),
"%s_%s range=%s:%d-%d 5'pad=%d 3'pad=%d "
"strand=%c repeatMasking=%s",
db,
name,
chrom, seqStart+1, seqEnd,
padding5, padding3,
(isRc ? '-' : '+'),
(maskRep ? repMasking : "none"));
faWriteNext(stdout, recName, cSeq->dna, cSeq->size);
freeDnaSeq(&cSeq);
}
void monitorSubmission(struct sqlConnection *conn)
/* Write out information about submission. */
{
char *url = trimSpaces(cgiString("url"));
cgiMakeHiddenVar("url", url);
struct edwSubmit *sub = edwMostRecentSubmission(conn, url);
time_t startTime = 0, endTime = 0, endUploadTime = 0;
if (sub == NULL)
{
int posInQueue = edwSubmitPositionInQueue(conn, url, NULL);
if (posInQueue == 0)
printf("%s is first in the submission queue, but upload has not started<BR>\n", url);
else if (posInQueue > 0)
printf("%s is in submission queue with %d submissions ahead of it<BR>\n", url, posInQueue);
else
{
printf("%s status unknown.", url);
}
}
else
{
startTime = sub->startUploadTime;
endUploadTime = sub->endUploadTime;
endTime = (endUploadTime ? endUploadTime : edwNow());
int timeSpan = endTime - startTime;
long long thisUploadSize = sub->byteCount - sub->oldBytes;
long long curSize = 0; // Amount of current file we know we've transferred.
/* Print title letting them know if upload is done or in progress. */
printf("<B>Submission by %s is ", userEmail);
if (!isEmpty(sub->errorMessage))
{
if (endUploadTime == 0)
printf("having problems...");
else
printf("stopped by uploader request.");
}
else if (endUploadTime != 0)
{
printf("uploaded.");
}
else
printf("in progress...");
printf("</B><BR>\n");
/* Print URL and how far along we are at the file level. */
if (!isEmpty(sub->errorMessage))
{
printf("<B>error:</B> %s<BR>\n", sub->errorMessage);
cgiMakeButton("getUrl", "try submission again");
printf("<BR>");
}
printf("<B>url:</B> %s<BR>\n", sub->url);
printf("<B>files count:</B> %d<BR>\n", sub->fileCount);
if (sub->oldFiles > 0)
printf("<B>files already in warehouse:</B> %u<BR>\n", sub->oldFiles);
if (sub->metaChangeCount > 0)
printf("<B>old files with new tags in this submission</B> %d<BR>", sub->metaChangeCount);
if (sub->oldFiles != sub->fileCount)
{
printf("<B>files transferred:</B> %u<BR>\n", sub->newFiles);
printf("<B>files remaining:</B> %u<BR>\n", sub->fileCount - sub->oldFiles - sub->newFiles);
}
/* Report validation status */
printf("<B>new files validated:</B> %u of %u<BR>\n", edwSubmitCountNewValid(sub, conn),
sub->newFiles);
/* Print error message, and in case of error skip file-in-transfer info. */
if (isEmpty(sub->errorMessage))
{
/* If possible print information about file en route */
if (endUploadTime == 0)
{
struct edwFile *ef = edwFileInProgress(conn, sub->id);
if (ef != NULL)
{
char path[PATH_LEN];
safef(path, sizeof(path), "%s%s", edwRootDir, ef->edwFileName);
if (ef->endUploadTime > 0)
curSize = ef->size;
else
curSize = paraFetchedSoFar(path);
printf("<B>file in route:</B> %s", ef->submitFileName);
printf(" (%d%% transferred)<BR>\n", (int)(100.0 * curSize / ef->size));
}
}
}
/* Report bytes transferred */
long long transferredThisTime = curSize + sub->newBytes;
printf("<B>total bytes transferred:</B> ");
long long totalTransferred = transferredThisTime + sub->oldBytes;
printLongWithCommas(stdout, totalTransferred);
printf(" of ");
printLongWithCommas(stdout, sub->byteCount);
if (sub->byteCount != 0)
printf(" (%d%%)<BR>\n", (int)(100.0 * totalTransferred / sub->byteCount));
else
printf("<BR>\n");
/* Report transfer speed if possible */
if (isEmpty(sub->errorMessage))
{
if (timeSpan > 0)
{
printf("<B>transfer speed:</B> ");
printLongWithCommas(stdout, (curSize + sub->newBytes)/timeSpan);
printf(" bytes/sec<BR>\n");
}
/* Report start time and duration */
printf("<B>submission started:</B> %s<BR>\n", ctime(&startTime));
struct dyString *duration = edwFormatDuration(timeSpan);
/* Try and give them an ETA if we aren't finished */
if (endUploadTime == 0 && timeSpan > 0)
{
printf("<B>time so far:</B> %s<BR>\n", duration->string);
double bytesPerSecond = (double)transferredThisTime/timeSpan;
long long bytesRemaining = thisUploadSize - curSize - sub->newBytes;
if (bytesPerSecond > 0)
{
long long estimatedFinish = bytesRemaining/bytesPerSecond;
struct dyString *eta = edwFormatDuration(estimatedFinish);
printf("<B>estimated finish in:</B> %s<BR>\n", eta->string);
}
}
else
{
printf("<B>submission time:</B> %s<BR>\n", duration->string);
cgiMakeButton("getUrl", "submit another data set");
}
}
}
cgiMakeButton("monitor", "refresh status");
if (endUploadTime == 0 && isEmpty(sub->errorMessage))
cgiMakeButton(stopButtonName, "stop upload");
printf(" <input type=\"button\" value=\"browse submissions\" "
"onclick=\"window.location.href='edwWebBrowse';\">\n");
edwPrintLogOutButton();
}
