Commits for angie
switch to files view, user index
v253_base to v254_preview (2011-06-28 to 2011-07-05) v254
- safef sticklerism.
- src/hg/hgTracks/chainTrack.c - lines changed 3, context: html, text, full: html, text
- src/hg/hgTracks/snakeTrack.c - lines changed 6, context: html, text, full: html, text
- Feature #986 (VCF+tabix support): basic custom-track how-to page.Lots of broken links -- still need to generate example files and to
write a configuration help page (hgVcfTrackHelp.html).
- src/hg/htdocs/goldenPath/help/vcf.html - lines changed 180, context: html, text, full: html, text
- MLQ #4463 (Problem with intersection in the table browser):an undefined Javascript variable was preventing drop-down menu changes from
triggering submission of hiddenForm. The primary table's inversion checkbox
is not drawn when the primary table is bigWig, so the onChange javascript
needs to omit that variable when primary table is bigWig.
- src/hg/hgTables/intersect.c - lines changed 1, context: html, text, full: html, text
- While debugging MLQ #4463, I noticed that jsMakeTrackingCheckBox wasimplemented in a way that would work if the checkbox was clicked a single
time (or odd number of times), but not if the checkbox was clicked an even
number of times. So if the user clicked the checkbox but changed their mind
and clicked again, and then changed the group or track, the checkbox would
change. Solution: make the variable toggle like the checkbox. Using +1 mod 2
so that our cart will see the expected 0 or 1 (not the logically inverted
true or false from ! negation).
- MLQ #4463 (problem w/intersection): another bug: canIntersect() wasreturning false for bigWig, so clicking submit on the intersection
page had no effect. Added clauses for bigWig and bigBed.
- src/hg/hgTables/intersect.c - lines changed 4, context: html, text, full: html, text
- While working on VCF, caught a couple little bugs in pgSnpTextRight:
1. When restoring the clipping rectangle after snapping a label to the
left, pgSnpTextRight thought it was using the track y offset -- but
the DrawAt routine only gets the item y, not track y. So when an item
on row N had a label that snapped left, the restored clipping
rectangle clips out any subsequent items on rows 0..N-1. The top
row(s) looks suspiciously empty, as if the packer hadn't fully
utilized the row, but it's only that those items are made invisible.
The fix was to get the real y clipping coords before clipping for
left text, so they could be restored correctly.
2. The text width was subtracted twice from the item pixel start before
the comparison with insideX (by the caller, and inside pgSnpTextRight).
Thus snapLeft was triggered too easily, causing overwriting of left
labels when items in the same row were close to each other and to the
left edge. Fix: the caller doesn't subtract (width + 2); pgSnpTextRight
uses textX, not x1, when drawing text.
- src/hg/hgTracks/simpleTracks.c - lines changed 12, context: html, text, full: html, text
- Feature #986 (VCF): When converting VCF to pgSnp for display, abbreviate long alleles.
- src/hg/hgTracks/vcfTrack.c - lines changed 25, context: html, text, full: html, text
- #986 (VCF): Loosening regexes for parsing VCF header to accomodate 1000 Genomes pilot release trio indel files.
- Feature #986 (VCF): keys can have <>'s, so html encode them for display.
switch to files view, user index