Total SQX 1.10 (packer plug-in)

[ghisler(Author)]

I think that it should be enabled by default. Why? If you want a very fast decompression of individual files, and size doesn't matter so much, you can as well use the standard ZIP compression (which I do). The real advantage of this special format mainly comes from the solid compression.

[Clo]

2ghisler(Author)

Good evening,

• This is not as so obvious as you tell above…
- I just tested a single *.BMP file, size 461.6 KB not compressed. I got :
- Internal ZIP —> 10.2 KB
- SQX normal —> 8.73 KB
- SQX solid —> 8.73 KB, exactly the same !
- SQX is configured to add recovery data by 3%.
- I guess that the "Solid" option is not needed by default for such a case, according to my tests (I performed a lot of others before too…)

 m.f.G.
Claude
Clo

[ghisler(Author)]

2Clo
Apparently you haven't understood the concept of solid compression! Try this: Make 10 copies of this image, and name them image01.bmp, image02.bmp .. image10.bmp. Now compress these 10 images once with normal and once with solid compression. You will be surprised!

To explain: Solid compression compresses multiple files together as if they were one single file, to detect and use similarities between these files. So if you compress a single file, you will not notice any difference...

[Clo]

2ghisler(Author)

If you want a very fast decompression of individual files,

- You told about individual files…

- I have translated the whole Help document, so I know how solid archives work, this is explained inside…

 m.f.G.
Claude
Clo

[Gral]

I think Mr. Ghisler told about decompression of individual file that means for example last file from archive contains 100 files. Time for this operation is the same as for decompression whole archive.

[Zenman]

2Gral

Yes, error occurs on WIN98, on WINXP i've not tested plugin yet.
I thought plugin works the same way on every 32-bit Windows

Yes it should work the same way but underlying system functions behave different sometimes.


2Zenman

My OS is winxp sp2, still get the same error message with the file "http://www.lefteous.de/tc/archives/totalsqx/totalsqx_1.0_translation.template.sqx"

Really strange. It works fine for me. The MD5 is
2f19ca8ddda33dcc224aa31fa56f5dad *totalsqx_1.0_translation.template.sqx


2both of you
Is it possible to pack files using sqx and unpack sqx archives or is it just this archive?

really strange!!!
with alt+shift+f9, I tested the file "totalsqx_supplimentary_package_1.01.sqx" wiith "no error"
unfortunately, with the file "totalsqx_1.0_translation.template.sqx" (md5=2F19CA8DDDA33DCC224AA31FA56F5DAD), it still shows 'archive is probably corrupted"
same error occurs with the file "xpdfsearch_1.02_source.sqx" (md5=B1D108018EC935BA15B84AF788AB91D0)
My os is win xp with sp2

[Gral]

Exactly the same on my Windows 98 SE.

[Lefteous]

2Zenman
2Gral
Thank you. I will try to find the reason.

[Clo]

Hello All !

¤ This plugin works fine, especially with regard to its newness… However, there is an issue remaining when one creates separate archives (one per selected file) about the decompression of these archives into separate sub-dirs. as one go, like it should.

- Currently, there is a name conflict when unpacking, because the plugin can't distinguish <file-name.xyz> as a file and
<file-name.xyz> as the directory to create, in which this file must be unpacked.

- I thought prior that the simplest could be that the plugin adds a mark, let's say an underscore “_” character to the name of the archive, keeping the extension as usual like :

<file-name.xyz_.sqx>

- Most certainly, there are no longer the “Acces denied” awful messages, but although the sub-dirs. are created alright, the files are upacked in the root, and the sub- dirs. stay empty

- Another way I tested (renaming the archives by hand, of course) is to create the archives as the file-name,
but without the extension.
- Whether one has several files having the same name and different extensions, a mark¦suffix could be added, allowing to get different archive names in the style :
<file-name.sqx> for the first file found,
<file-name_01.sqx> for the second file, same name, different extension and so on.
- This allows up to 100 different extensions for the same file name; in the practice, this might be largely sufficient…
- In that way, all files are alright unpacked, each into its own directory,
like you can see by the shown sizes on this real screen shot.

• Hoping this could help,

Kind regards,
Claude
Clo

[Vansumsen]

- I tested the two systems proposed by Clo. I can confirm :
1. Adding just an underscore is not satisfactory (same result as Clo's one)
2. The second way without the extension, and adding a marking for the duplicates works just fine…

I thought to a script in order to make that renaming «automatically» temporarily, but it seems pretty complicated…

Regards

Paul

[Lefteous]

2Clo
I will investigate this problem. It seems it's related to the rename on file name conflicts problem mentioned in the documentation.
But it could be also related to another problem which should be fixed in the next version.

[Lefteous]

2everyone having the unpack problem on some files

The problem is a misleading error message. Instead of a CRC or bad archive error an out of memory error should be displayed.
I have uploaded a less memory-hungry archive to my site:
http://www.lefteous.de/tc/archives/totalsqx/totalsqx_1.0_translation.template.sqx
I will add this problem to the known bugs section until its fixed.

A related problem in the current version is in the packer configuration dialog which creates the impression that the text order and text memory options are automatically set by the text compression option but this is not true.
This bug will be fixed in the next version.

[Zenman]

2everyone having the unpack problem on some files

The problem is a misleading error message. Instead of a CRC or bad archive error an out of memory error should be displayed.
I have uploaded a less memory-hungry archive to my site:
http://www.lefteous.de/tc/archives/totalsqx/totalsqx_1.0_translation.template.sqx
I will add this problem to the known bugs section until its fixed.

A related problem in the current version is in the packer configuration dialog which creates the impression that the text order and text memory options are automatically set by the text compression option but this is not true.
This bug will be fixed in the next version.

This time the unpacking is OK
btw, is there some options to tweak the 'unpacking' memory usage when 'sqeezing' files/dirs?

[Lefteous]

2Zenman

s there some options to tweak the 'unpacking' memory usage when 'sqeezing' files/dirs?

I think the most important setting here is the text compression setting. 32 MB is a good value.
Another setting to be considered is the dictonary size. Setting it to 4 MB should be OK for low memory systems.

The current version contains a bug which makes it impossible to set the text compression options. This bug will be fixed in the upcoming update.
http://www.ghisler.ch/board/viewtopic.php?p=73733#73733

[petermad]

2Lefteous

A Danish translation of this excellent plugin can be downloaded at http://madsenworld.dk/tcmd/totalsqx_danish100.sqx

Sorry I didn't use the .ods file - I don't have OpenOffice installed.

There seem to be a discrepancy in the last line of the main language files in English versus German:

Code: Select all

10074=Cannot create a self-extracting archive from a multi-volume archive.

Code: Select all

10074=Kann kein mehrteiliges Archiv aus einem selbstextrahierenden Archiv erstellen.

I guess the English text is the correct ?