mirror of
https://github.com/flynx/ImageGrid.git
synced 2025-10-29 18:30:09 +00:00
made filename sort aware of image sequence overflow...
Signed-off-by: Alex A. Naanou <alex.nanou@gmail.com>
This commit is contained in:
parent
b822818555
commit
b06c5f05f8
90
ui/data.js
90
ui/data.js
@ -1782,11 +1782,99 @@ function sortImages(cmp, reverse){
|
||||
function sortImagesByDate(reverse){
|
||||
return sortImages(reverse)
|
||||
}
|
||||
function sortImagesByName(reverse){
|
||||
function sortImagesByFileName(reverse){
|
||||
return sortImages(imageNameCmp, reverse)
|
||||
}
|
||||
|
||||
|
||||
// Sort images while taking into account name sequence overflows
|
||||
//
|
||||
// A name sequence overflow is when file name sequence overflows over
|
||||
// *9999 and then resets to *0001...
|
||||
//
|
||||
// For this to be applicable:
|
||||
// - filenames must comply with /....[0-9]{4}/
|
||||
// - the total number of files in sequence is < 10K
|
||||
// XXX a simplification...
|
||||
// there could be more than 10K images but then we will need to
|
||||
// either take dates or folder names into account...
|
||||
// - the lowest filename in set must be near seq 0001
|
||||
// - the highest filename in set must be near seq 9999
|
||||
// - there must be a gap somewhere in the set
|
||||
// this gap size is roughly close to 10K - N where N is the total
|
||||
// number of files in set
|
||||
// XXX a simplification...
|
||||
//
|
||||
// NOTE: this will cut at the largest gap between sequence numbers.
|
||||
//
|
||||
// XXX it's also a good idea to write an image serial number sort...
|
||||
// XXX is this overcomplicated???
|
||||
function sortImagesByNameWithSeqOverflow(reverse, proximity){
|
||||
proximity = proximity == null ? 10 : proximity
|
||||
|
||||
// sort and check names...
|
||||
// NOTE: we do not usually have a filename seq 0000...
|
||||
if(DATA.order.length < 9999){
|
||||
var pattern = /....[0-9]{4}[a-z]?(\....[.])?/
|
||||
var need_to_fix = true
|
||||
|
||||
function cmp(a, b){
|
||||
// check filename compliance...
|
||||
var aa = pattern.test(IMAGES[a].path.split('/').pop())
|
||||
var bb = pattern.test(IMAGES[b].path.split('/').pop())
|
||||
if(need_to_fix && !aa && !bb){
|
||||
need_to_fix = false
|
||||
}
|
||||
return imageNameCmp(a, b)
|
||||
}
|
||||
function getSeq(gid){
|
||||
return parseInt(IMAGES[gid].path.split('/').pop().slice(4))
|
||||
}
|
||||
|
||||
// revert to normal sort my name...
|
||||
} else {
|
||||
return sortImagesByName(reverse)
|
||||
}
|
||||
|
||||
DATA.order.sort(cmp)
|
||||
|
||||
if(need_to_fix
|
||||
// check if first and last are close to 0001 and 9999 resp.
|
||||
&& getSeq(DATA.order[0]) <= proximity
|
||||
&& getSeq(DATA.order[DATA.order.length-1]) >= 9999-proximity){
|
||||
|
||||
console.log('fixing...')
|
||||
|
||||
// find the largest gap position...
|
||||
var pos = null
|
||||
var gap = 0
|
||||
for(var i=1; i<DATA.order.length; i++){
|
||||
var n_gap = Math.max(getSeq(DATA.order[i])-getSeq(DATA.order[i-1]), gap)
|
||||
if(n_gap != gap){
|
||||
pos = i
|
||||
gap = n_gap
|
||||
}
|
||||
}
|
||||
|
||||
// splice...
|
||||
if(gap > proximity){
|
||||
DATA.order = DATA.order.splice(pos).concat(DATA.order)
|
||||
}
|
||||
}
|
||||
|
||||
if(reverse){
|
||||
DATA.order.reverse()
|
||||
}
|
||||
|
||||
updateRibbonOrder()
|
||||
}
|
||||
|
||||
|
||||
// Set the default filename sort...
|
||||
// XXX is this the correct way to do this???
|
||||
var sortImagesByName = sortImagesByNameWithSeqOverflow
|
||||
|
||||
|
||||
|
||||
/************************************************** Manual sorting ***/
|
||||
|
||||
|
||||
Loading…
x
Reference in New Issue
Block a user