f587148862bf76eb764a550db5d52b00318679fc
angie
  Wed May 29 10:28:10 2019 -0700
Report item counts correctly for big* types whose files are specified in db tables instead of bigDataUrl settings.  refs #18869, #23521

diff --git src/hg/hubApi/list.c src/hg/hubApi/list.c
index 1db5129..e41e290 100644
--- src/hg/hubApi/list.c
+++ src/hg/hubApi/list.c
@@ -1,623 +1,652 @@
 /* manage endpoint /list/ functions */
 
 #include "dataApi.h"
 
 static void hubPublicJsonData(struct jsonWrite *jw, struct hubPublic *el,
   int columnCount, char **columnNames)
 /* Print array data for one row from hubPublic table, order here
  * must be same as was stated in the columnName header element
  * This code should be in hg/lib/hubPublic.c (which does not exist)
  */
 {
 int i = 0;
 jsonWriteObjectStart(jw, NULL);
 jsonWriteString(jw, columnNames[i++], el->hubUrl);
 jsonWriteString(jw, columnNames[i++], el->shortLabel);
 jsonWriteString(jw, columnNames[i++], el->longLabel);
 jsonWriteString(jw, columnNames[i++], el->registrationTime);
 jsonWriteNumber(jw, columnNames[i++], (long long)el->dbCount);
 jsonWriteString(jw, columnNames[i++], el->dbList);
 jsonWriteString(jw, columnNames[i++], el->descriptionUrl);
 jsonWriteObjectEnd(jw);
 }
 
 static void jsonPublicHubs()
 /* output the hubPublic SQL table */
 {
 char *extraArgs = verifyLegalArgs(argListPublicHubs); /* no extras allowed */
 if (extraArgs)
     apiErrAbort(err400, err400Msg, "extraneous arguments found for function /list/publicHubs '%s'", extraArgs);
 
 struct sqlConnection *conn = hConnectCentral();
 char *dataTime = sqlTableUpdate(conn, hubPublicTableName());
 time_t dataTimeStamp = sqlDateToUnixTime(dataTime);
 replaceChar(dataTime, ' ', 'T');	/* ISO 8601 */
 struct hubPublic *el = hubPublicDbLoadAll();
 struct jsonWrite *jw = apiStartOutput();
 jsonWriteString(jw, "dataTime", dataTime);
 jsonWriteNumber(jw, "dataTimeStamp", (long long)dataTimeStamp);
 freeMem(dataTime);
 // redundant: jsonWriteString(jw, "tableName", hubPublicTableName());
 char **columnNames = NULL;
 char **columnTypes = NULL;
 int *jsonTypes = NULL;
 int columnCount = tableColumns(conn, jw, hubPublicTableName(), &columnNames,
     &columnTypes, &jsonTypes);
 jsonWriteListStart(jw, "publicHubs");
 for ( ; el != NULL; el = el->next )
     {
     hubPublicJsonData(jw, el, columnCount, columnNames);
     }
 jsonWriteListEnd(jw);
 apiFinishOutput(0, NULL, jw);
 hDisconnectCentral(&conn);
 }
 
 static void dbDbJsonData(struct jsonWrite *jw, struct dbDb *el,
     int columnCount, char **columnNames)
 /* Print out dbDb table element in JSON format.
  * must be same as was stated in the columnName header element
  * This code should be over in hg/lib/dbDb.c
  */
 {
 int i = 0;
 jsonWriteObjectStart(jw, el->name);
 i++;
 // redundant: jsonWriteString(jw, NULL, el->name);
 jsonWriteString(jw, columnNames[i++], el->description);
 jsonWriteString(jw, columnNames[i++], el->nibPath);
 jsonWriteString(jw, columnNames[i++], el->organism);
 jsonWriteString(jw, columnNames[i++], el->defaultPos);
 jsonWriteNumber(jw, columnNames[i++], (long long)el->active);
 jsonWriteNumber(jw, columnNames[i++], (long long)el->orderKey);
 jsonWriteString(jw, columnNames[i++], el->genome);
 jsonWriteString(jw, columnNames[i++], el->scientificName);
 jsonWriteString(jw, columnNames[i++], el->htmlPath);
 jsonWriteNumber(jw, columnNames[i++], (long long)el->hgNearOk);
 jsonWriteNumber(jw, columnNames[i++], (long long)el->hgPbOk);
 jsonWriteString(jw, columnNames[i++], el->sourceName);
 jsonWriteNumber(jw, columnNames[i++], (long long)el->taxId);
 jsonWriteObjectEnd(jw);
 }
 
 static void jsonDbDb()
 /* output the dbDb SQL table */
 {
 char *extraArgs = verifyLegalArgs(argListUcscGenomes); /* no extras allowed */
 if (extraArgs)
     apiErrAbort(err400, err400Msg, "extraneous arguments found for function /list/ucscGenomes '%s'", extraArgs);
 
 struct sqlConnection *conn = hConnectCentral();
 char *dataTime = sqlTableUpdate(conn, "dbDb");
 time_t dataTimeStamp = sqlDateToUnixTime(dataTime);
 replaceChar(dataTime, ' ', 'T');	/* ISO 8601 */
 struct dbDb *dbList = ucscDbDb();
 struct dbDb *el;
 struct jsonWrite *jw = apiStartOutput();
 jsonWriteString(jw, "dataTime", dataTime);
 jsonWriteNumber(jw, "dataTimeStamp", (long long)dataTimeStamp);
 freeMem(dataTime);
 char **columnNames = NULL;
 char **columnTypes = NULL;
 int *jsonTypes = NULL;
 int columnCount = tableColumns(conn, jw, "dbDb", &columnNames, &columnTypes,
     &jsonTypes);
 jsonWriteObjectStart(jw, "ucscGenomes");
 for ( el=dbList; el != NULL; el = el->next )
     {
     dbDbJsonData(jw, el, columnCount, columnNames);
     }
 jsonWriteObjectEnd(jw);
 apiFinishOutput(0, NULL, jw);
 hDisconnectCentral(&conn);
 }
 
 static void hubChromInfoJsonOutput(FILE *f, char *hubUrl, char *genome)
 /* for given hubUrl list the chromosomes in the sequence for specified genome
  */
 {
 struct trackHub *hub = errCatchTrackHubOpen(hubUrl);
 struct trackHubGenome *ge = NULL;
 char *track = cgiOptionalString("track");
 
 if (isEmpty(genome))
     apiErrAbort(err400, err400Msg, "must specify a 'genome=name' with hubUrl for endpoint: /list/chromosomes?hubUrl=%s;genome=<empty>", hubUrl);
 
 struct trackHubGenome *foundGenome = NULL;
 
 for (ge = hub->genomeList; ge; ge = ge->next)
     {
     if (sameOk(genome, ge->name))
 	{
 	foundGenome = ge;
 	continue;	/* found genome */
 	}
     }
 
 if (NULL == foundGenome)
     apiErrAbort(err400, err400Msg, "can not find specified 'genome=%s' for endpoint: /list/chromosomes?hubUrl=%s;genome=%s", genome, hubUrl, genome);
 
 struct jsonWrite *jw = apiStartOutput();
 jsonWriteString(jw, "hubUrl", hubUrl);
 jsonWriteString(jw, "genome", genome);
 if (isNotEmpty(track))
     {
     jsonWriteString(jw, "track", track);
     struct trackDb *tdb = obtainTdb(foundGenome, NULL);
     if (NULL == tdb)
 	apiErrAbort(err400, err400Msg, "failed to find a track hub definition in genome=%s for endpoint '/list/chromosomes' given hubUrl=%s'", genome, hubUrl);
 
     struct trackDb *thisTrack = findTrackDb(track, tdb);
     if (NULL == thisTrack)
 	apiErrAbort(err400, err400Msg, "failed to find specified track=%s in genome=%s for endpoint '/list/chromosomes'  given hubUrl='%s'", track, genome, hubUrl);
 
     char *bigDataUrl = trackDbSetting(thisTrack, "bigDataUrl");
     struct bbiFile *bbi = bigFileOpen(thisTrack->type, bigDataUrl);
     struct bbiChromInfo *chrList = bbiChromList(bbi);
     slSort(chrList, chromInfoCmp);
     struct bbiChromInfo *el = chrList;
     jsonWriteNumber(jw, "chromCount", (long long)slCount(chrList));
     jsonWriteObjectStart(jw, "chromosomes");
     for ( ; el; el = el->next )
 	{
 	jsonWriteNumber(jw, el->name, (long long)el->size);
 	}
     jsonWriteObjectEnd(jw);	/* chromosomes */
     }
 else
     {
     struct chromInfo *ci = NULL;
     /* might be a track hub on a UCSC database */
     if (isEmpty(foundGenome->twoBitPath))
 	{
 	struct sqlConnection *conn = hAllocConnMaybe(foundGenome->trackHub->defaultDb);
 	if (NULL == conn)
 	    apiErrAbort(err400, err400Msg, "can not find 'genome=%s' for endpoint '/list/chromosomes", foundGenome->trackHub->defaultDb);
 	else
 	    hFreeConn(&conn);
 	ci = createChromInfoList(NULL, foundGenome->trackHub->defaultDb);
 	}
     else
 	{
 	ci = trackHubAllChromInfo(foundGenome->name);
 	}
     slSort(ci, chromInfoCmp);
     jsonWriteNumber(jw, "chromCount", (long long)slCount(ci));
     jsonWriteObjectStart(jw, "chromosomes");
     struct chromInfo *el = ci;
     for ( ; el != NULL; el = el->next )
 	{
 	jsonWriteNumber(jw, el->chrom, (long long)el->size);
 	}
     jsonWriteObjectEnd(jw);	/* chromosomes */
     }
 apiFinishOutput(0, NULL, jw);
 }
 
 static char *validChromName(struct sqlConnection *conn, char *db, char *table,
    char **splitTableName, struct hTableInfo **tableInfo)
 /* determine what the 'chrom' name should be for this table (aka track)
  * this function could be used in getData() also TBD
  */
 {
 static char *returnChrom = NULL;
 /* to be determined if this table name is used or is some other name */
 char *sqlTableName = cloneString(table);
 
 /* 'track' name in trackDb usually refers to a SQL 'table' */
 struct trackDb *tdb = obtainTdb(NULL, db);
 struct trackDb *thisTrack = findTrackDb(table,tdb);
 /* however, the trackDb might have a specific table defined instead */
 char *tableName = trackDbSetting(thisTrack, "table");
 if (isNotEmpty(tableName))
     {
     freeMem(sqlTableName);
     sqlTableName = cloneString(tableName);
     }
 
 /* this function knows how to deal with split chromosomes, the NULL
  * here for the chrom name means to use the first chrom name in chromInfo
  */
 struct hTableInfo *hti = hFindTableInfoWithConn(conn, NULL, sqlTableName);
 *tableInfo = hti;	/* returning to caller */
 /* check if table name needs to be modified */
 if (hti && hti->isSplit)
     {
     char *defaultChrom = hDefaultChrom(db);
     char fullTableName[256];
     safef(fullTableName, sizeof(fullTableName), "%s_%s", defaultChrom, hti->rootName);
     freeMem(sqlTableName);
     sqlTableName = cloneString(fullTableName);
     *splitTableName = cloneString(fullTableName);	/* return to caller */
     }
 else
     {
     *splitTableName = sqlTableName;	/* return to caller */
     }
 
 /* may need to extend this in the future for other track types */
 if (sqlColumnExists(conn, sqlTableName, "chrom"))	/* standard bed tables */
     returnChrom = cloneString("chrom");
 else if (sqlColumnExists(conn, sqlTableName, "tName"))	/* track type psl */
     returnChrom = cloneString("tName");
 else if (sqlColumnExists(conn, sqlTableName, "genoName"))	/* track type rmsk */
     returnChrom = cloneString("genoName");
 
 return returnChrom;
 }
 
 static void chromInfoJsonOutput(FILE *f, char *db)
 /* for given db, if there is a track, list the chromosomes in that track,
  * for no track, simply list the chromosomes in the sequence
  */
 {
 char *splitSqlTable = NULL;
 struct hTableInfo *tableInfo = NULL;
 char *chromName = NULL;
 char *table = cgiOptionalString("track");
 struct sqlConnection *conn = hAllocConnMaybe(db);
 if (NULL == conn)
     apiErrAbort(err400, err400Msg, "can not find 'genome=%s' for endpoint '/list/chromosomes", db);
 
 if (table)
     chromName = validChromName(conn, db, table, &splitSqlTable, &tableInfo);
 
 /* in trackDb language: track == table */
 /* punting on split tables, just return chromInfo */
 if (table && chromName && ! (tableInfo && tableInfo->isSplit) )
     {
     if (! sqlTableExists(conn, splitSqlTable))
 	apiErrAbort(err400, err400Msg, "can not find specified 'track=%s' for endpoint: /list/chromosomes?genome=%s;track=%s", table, db, table);
 
     if (sqlColumnExists(conn, splitSqlTable, chromName))
 	{
 	char *dataTime = sqlTableUpdate(conn, splitSqlTable);
 	time_t dataTimeStamp = sqlDateToUnixTime(dataTime);
 	replaceChar(dataTime, ' ', 'T');	/* ISO 8601 */
         struct jsonWrite *jw = apiStartOutput();
 	jsonWriteString(jw, "genome", db);
 	jsonWriteString(jw, "track", table);
 	jsonWriteString(jw, "dataTime", dataTime);
 	jsonWriteNumber(jw, "dataTimeStamp", (long long)dataTimeStamp);
 	freeMem(dataTime);
         struct slPair *list = NULL;
 	char query[2048];
         sqlSafef(query, sizeof(query), "select distinct %s from %s", chromName, splitSqlTable);
 	struct sqlResult *sr = sqlGetResult(conn, query);
 	char **row;
 	while ((row = sqlNextRow(sr)) != NULL)
     	{
             int size = hChromSize(db, row[0]);
 	    slAddHead(&list, slPairNew(row[0], intToPt(size)));
     	}
 	sqlFreeResult(&sr);
         slPairIntSort(&list);
         slReverse(&list);
         jsonWriteNumber(jw, "chromCount", (long long)slCount(list));
 	jsonWriteObjectStart(jw, "chromosomes");
         struct slPair *el = list;
         for ( ; el != NULL; el = el->next )
             jsonWriteNumber(jw, el->name, (long long)ptToInt(el->val));
 	jsonWriteObjectEnd(jw);	/* chromosomes */
 	apiFinishOutput(0, NULL, jw);
 	}
     else
 	apiErrAbort(err400, err400Msg, "track '%s' is not a position track, request table without chrom specification, genome: '%s'", table, db);
     }
 else if (table && !chromName)	/* only allowing position tables at this time */
 	apiErrAbort(err400, err400Msg, "track '%s' is not a position track, request table without chrom specification, genome: '%s'", table, db);
 else
     {
     char *dataTime = sqlTableUpdate(conn, "chromInfo");
     time_t dataTimeStamp = sqlDateToUnixTime(dataTime);
     replaceChar(dataTime, ' ', 'T');	/* ISO 8601 */
     struct chromInfo *ciList = createChromInfoList(NULL, db);
     slSort(ciList, chromInfoCmp);
     struct chromInfo *el = ciList;
     struct jsonWrite *jw = apiStartOutput();
     jsonWriteString(jw, "genome", db);
     jsonWriteString(jw, "dataTime", dataTime);
     if (tableInfo && tableInfo->isSplit)	/* the split table punt */
 	jsonWriteString(jw, "track", table);
     jsonWriteNumber(jw, "dataTimeStamp", (long long)dataTimeStamp);
     freeMem(dataTime);
     jsonWriteNumber(jw, "chromCount", (long long)slCount(ciList));
     jsonWriteObjectStart(jw, "chromosomes");
     for ( ; el != NULL; el = el->next )
 	{
         jsonWriteNumber(jw, el->chrom, (long long)el->size);
 	}
     jsonWriteObjectEnd(jw);	/* chromosomes */
     apiFinishOutput(0, NULL, jw);
     }
 hFreeConn(&conn);
 }
 
 static long long bbiItemCount(char *bigDataUrl, char *type)
 /* check the bigDataUrl to see what the itemCount is there */
 {
 long long itemCount = 0;
 struct errCatch *errCatch = errCatchNew();
 if (errCatchStart(errCatch))
     {
     if (allowedBigBedType(type))
         {
         struct bbiFile *bbi = NULL;
         bbi = bigBedFileOpen(bigDataUrl);
         itemCount = bigBedItemCount(bbi);
         bbiFileClose(&bbi);
         }
     else if (startsWithWord("bigWig", type))
         {
         struct bbiFile *bwf = bigWigFileOpen(bigDataUrl);
         struct bbiSummaryElement sum = bbiTotalSummary(bwf);
         itemCount = sum.validCount;
         bbiFileClose(&bwf);
         }
+    // NOTE: htslib does not appear to have a function to extract a count from BAI or TBI (tabix)
+    // indexes.  One could probably be added by adding code similar to samtools' idxstats subcommand,
+    // But in the meantime, this is not supported for types vcfTabix and bam.
     }
 errCatchEnd(errCatch);
 errCatchFree(&errCatch);
 return itemCount;
 }
 
+static long long bbiTableItemCount(struct sqlConnection *conn, char *type, char *tableName)
+/* Given a tableName that has a fileName column pointing to big*, bam or vcfTabix files, return the
+ * total itemCount from all rows (BAM and VCF tables may have one row per chrom). */
+{
+long long itemCount = 0;
+char query[2048];
+sqlSafef(query, sizeof query, "select fileName from %s", tableName);
+struct sqlResult *sr = sqlGetResult(conn, query);
+char **row;
+while ((row = sqlNextRow(sr)) != NULL)
+    {
+    itemCount += bbiItemCount(row[0], type);
+    }
+sqlFreeResult(&sr);
+return itemCount;
+}
+
 static long long dataItemCount(char *db, struct trackDb *tdb)
 /* determine how many items are in this data set */
 {
 boolean isContainer = tdbIsComposite(tdb) || tdbIsCompositeView(tdb);
 if (trackDbSetting(tdb, "container"))
     isContainer = TRUE;
 long long itemCount = 0;
 if (isContainer)	/* containers have no data items */
     return itemCount;
 if (sameWord("downloadsOnly", tdb->type))
     return itemCount;
 
 char *bigDataUrl = trackDbSetting(tdb, "bigDataUrl");
 if (isNotEmpty(bigDataUrl))
     itemCount = bbiItemCount(bigDataUrl, tdb->type);
 else
     {
     /* prepare for getting table row count, find table name */
     /* the trackDb might have a specific table defined */
     char *tableName = trackDbSetting(tdb, "table");
     if (isEmpty(tableName))
 	tableName = trackDbSetting(tdb, "track");
     if (isNotEmpty(tableName))
 	{
 	struct sqlConnection *conn = hAllocConnMaybe(db);
 	if (conn)
 	    {
+            if ((startsWith("big", tdb->type) ||
+                 sameString("vcfTabix", tdb->type) || sameString("bam", tdb->type)) &&
+                sqlColumnExists(conn, tableName, "fileName"))
+                {
+                itemCount = bbiTableItemCount(conn, tdb->type, tableName);
+                }
+            else
+                {
                 /* punting on split tables, return zero */
                 struct hTableInfo *hti =
                     hFindTableInfoWithConn(conn, NULL, tableName);
                 if (hti && hti->isSplit)
                     {
                     itemCount = 0;
                     }
                 else
                     {
                     char query[2048];
                     sqlSafef(query, sizeof(query), "select count(*) from %s", tableName);
                     itemCount = sqlQuickNum(conn, query);
                     }
+                }
             hFreeConn(&conn);
 	    }
 	}
     }
 return itemCount;
 }
 
 static void recursiveTrackList(struct jsonWrite *jw, struct trackDb *tdb,
     char *db)
 /* output trackDb tags only for real tracks, not containers,
  * recursive when subtracks exist
  */
 {
 boolean isContainer = tdbIsComposite(tdb) || tdbIsCompositeView(tdb);
 
 /* do *NOT* print containers when 'trackLeavesOnly' requested */
 if (! (trackLeavesOnly && isContainer) )
     {
     boolean protectedData = FALSE;
     long long itemCount = 0;
     if (trackDbSetting(tdb, "tableBrowser"))
 	protectedData = TRUE;
     else
 	itemCount = dataItemCount(db, tdb);
     jsonWriteObjectStart(jw, tdb->track);
     if (tdbIsComposite(tdb))
         jsonWriteString(jw, "compositeContainer", "TRUE");
     if (tdbIsCompositeView(tdb))
         jsonWriteString(jw, "compositeViewContainer", "TRUE");
     jsonWriteString(jw, "shortLabel", tdb->shortLabel);
     jsonWriteString(jw, "type", tdb->type);
     jsonWriteString(jw, "longLabel", tdb->longLabel);
     jsonWriteNumber(jw, "itemCount", itemCount);
     if (tdb->parent)
         {
         jsonWriteString(jw, "parent", tdb->parent->track);
 	if (tdb->parent->parent)
 	    jsonWriteString(jw, "parentParent", tdb->parent->parent->track);
         }
     if (tdb->settingsHash)
         {
         struct hashEl *hel;
         struct hashCookie hc = hashFirst(tdb->settingsHash);
         while ((hel = hashNext(&hc)) != NULL)
             {
             if (sameWord("track", hel->name))
 		continue;	// already output in header
             if (sameWord("tableBrowser", hel->name))
 		jsonWriteBoolean(jw, "protectedData", TRUE);
             else if (isEmpty((char *)hel->val))
 		jsonWriteString(jw, hel->name, "empty");
             else if (protectedData && sameWord(hel->name, "bigDataUrl"))
 		jsonWriteString(jw, hel->name, "protectedData");
 	    else
 		jsonWriteString(jw, hel->name, (char *)hel->val);
             }
         }
 
     if (tdb->subtracks)
 	{
 	struct trackDb *el = NULL;
 	for (el = tdb->subtracks; el != NULL; el = el->next )
 	    recursiveTrackList(jw, el, db);
 	}
 
     jsonWriteObjectEnd(jw);
     }
 else if (tdb->subtracks)
     {
     struct trackDb *el = NULL;
     for (el = tdb->subtracks; el != NULL; el = el->next )
 	recursiveTrackList(jw, el, db);
     }
 }	/*	static void recursiveTrackList()	*/
 
 static void trackDbJsonOutput(char *db, FILE *f)
 /* return track list from specified UCSC database name */
 {
 struct sqlConnection *conn = hAllocConnMaybe(db);
 if (NULL == conn)
     apiErrAbort(err400, err400Msg, "can not find 'genome=%s' for endpoint '/list/tracks", db);
 
 char *dataTime = sqlTableUpdate(conn, "trackDb");
 time_t dataTimeStamp = sqlDateToUnixTime(dataTime);
 replaceChar(dataTime, ' ', 'T');	/* ISO 8601 */
 hFreeConn(&conn);
 struct trackDb *tdbList = obtainTdb(NULL, db);
 struct jsonWrite *jw = apiStartOutput();
 jsonWriteString(jw, "genome", db);
 jsonWriteString(jw, "dataTime", dataTime);
 jsonWriteNumber(jw, "dataTimeStamp", (long long)dataTimeStamp);
 freeMem(dataTime);
 struct trackDb *el = NULL;
 for (el = tdbList; el != NULL; el = el->next )
     {
     recursiveTrackList(jw, el, db);
     }
 apiFinishOutput(0, NULL, jw);
 }	/*	static void trackDbJsonOutput(char *db, FILE *f)	*/
 
 void apiList(char *words[MAX_PATH_INFO])
 /* 'list' function words[1] is the subCommand */
 {
 if (sameWord("publicHubs", words[1]))
     jsonPublicHubs();
 else if (sameWord("ucscGenomes", words[1]))
     jsonDbDb();
 else if (sameWord("hubGenomes", words[1]))
     {
     char *extraArgs = verifyLegalArgs(argListHubGenomes); /* only one allowed */
     if (extraArgs)
 	apiErrAbort(err400, err400Msg, "extraneous arguments found for function /list/hubGenomes '%s'", extraArgs);
 
     char *hubUrl = cgiOptionalString("hubUrl");
     if (isEmpty(hubUrl))
 	apiErrAbort(err400, err400Msg, "must supply hubUrl='http:...' some URL to a hub for /list/hubGenomes");
 
     struct trackHub *hub = errCatchTrackHubOpen(hubUrl);
     if (hub->genomeList)
 	{
 	slNameSort((struct slName **)&hub->genomeList);
         struct jsonWrite *jw = apiStartOutput();
 	jsonWriteString(jw, "hubUrl", hubUrl);
         jsonWriteObjectStart(jw, "genomes");
 	struct trackHubGenome *el;
         for ( el = hub->genomeList; el; el = el->next)
 	    {
 	    jsonWriteObjectStart(jw, el->name);
 	    jsonWriteString(jw, "organism", el->organism);
 	    jsonWriteString(jw, "description", el->description);
 	    jsonWriteString(jw, "trackDbFile", el->trackDbFile);
 	    jsonWriteString(jw, "twoBitPath", el->twoBitPath);
 	    jsonWriteString(jw, "groups", el->groups);
 	    jsonWriteString(jw, "defaultPos", el->defaultPos);
 	    jsonWriteNumber(jw, "orderKey", el->orderKey);
 	    jsonWriteObjectEnd(jw);
 	    }
 	jsonWriteObjectEnd(jw);
 	apiFinishOutput(0, NULL, jw);
 	}
     }
 else if (sameWord("tracks", words[1]))
     {
     char *extraArgs = verifyLegalArgs(argListTracks);
     if (extraArgs)
 	apiErrAbort(err400, err400Msg, "extraneous arguments found for function /list/tracks '%s'", extraArgs);
 
     char *hubUrl = cgiOptionalString("hubUrl");
     char *genome = cgiOptionalString("genome");
     char *db = cgiOptionalString("genome");
     if (isEmpty(hubUrl) && isNotEmpty(db))
 	{
 	struct sqlConnection *conn = hAllocConnMaybe(db);
         if (NULL == conn)
 	    apiErrAbort(err400, err400Msg, "can not find 'genome=%s' for endpoint '/list/tracks", db);
 	else
 	    hFreeConn(&conn);
 	}
     if (isEmpty(hubUrl) && isEmpty(db))
       apiErrAbort(err400, err400Msg, "missing hubUrl or genome name for endpoint /list/tracks");
     if (isEmpty(hubUrl))	// missing hubUrl implies UCSC database
 	{
         trackDbJsonOutput(db, stdout);	// only need db for this function
 	return;
 	}
     if (isEmpty(genome) || isEmpty(hubUrl))
 	{
         if (isEmpty(genome))
 	    apiErrAbort(err400, err400Msg, "must supply genome='someName' the name of a genome in a hub for /list/tracks\n");
 	if (isEmpty(hubUrl))
             apiErrAbort(err400, err400Msg, "must supply hubUrl='http:...' some URL to a hub for /list/tracks");
 	}
     struct trackHub *hub = errCatchTrackHubOpen(hubUrl);
     struct trackHubGenome *hubGenome = findHubGenome(hub, genome,
 	"/list/tracks", hubUrl);
     struct trackDb *tdbList = obtainTdb(hubGenome, NULL);
     struct jsonWrite *jw = apiStartOutput();
     jsonWriteString(jw, "hubUrl", hubUrl);
     jsonWriteObjectStart(jw, hubGenome->name);
     struct trackDb *el = NULL;
     for (el = tdbList; el != NULL; el = el->next )
 	    {
 	    recursiveTrackList(jw, el, db);
 	    }
     jsonWriteObjectEnd(jw);
     apiFinishOutput(0, NULL, jw);
     }
 else if (sameWord("chromosomes", words[1]))
     {
     char *extraArgs = verifyLegalArgs(argListChromosomes);
     if (extraArgs)
 	apiErrAbort(err400, err400Msg, "extraneous arguments found for function /list/chromosomes '%s'", extraArgs);
 
     char *hubUrl = cgiOptionalString("hubUrl");
     char *genome = cgiOptionalString("genome");
     char *db = cgiOptionalString("genome");
     if (isEmpty(hubUrl) && isNotEmpty(db))
 	{
 	struct sqlConnection *conn = hAllocConnMaybe(db);
         if (NULL == conn)
 	    apiErrAbort(err400, err400Msg, "can not find 'genome=%s' for endpoint '/list/chromosomes", db);
 	else
 	    hFreeConn(&conn);
 	}
     if (isEmpty(hubUrl) && isEmpty(db))
         apiErrAbort(err400, err400Msg, "must supply hubUrl or genome name for endpoint '/list/chromosomes", hubUrl, db);
 
     if (isEmpty(hubUrl))	// missing hubUrl implies UCSC database
 	{
         chromInfoJsonOutput(stdout, db);
 	return;
 	}
     else
 	{
         hubChromInfoJsonOutput(stdout, hubUrl, genome);
 	return;
 	}
     }
 else
     apiErrAbort(err400, err400Msg, "do not recognize endpoint function: '/%s/%s'", words[0], words[1]);
 }	/*	void apiList(char *words[MAX_PATH_INFO])        */