Bug 36819

I tried to look at the db data before and after the change, and I don't see a difference. I thought maybe it was NULL and after it was filled with the correct values, but the values are already there...

Layout 17 before step 13

MariaDB [koha_kohadev]> select * from creator_layouts where layout_id = 17 \G;
*************************** 1. row ***************************
    layout_id: 17
 barcode_type: CODE39
  start_label: 1
printing_type: BIBBAR
  layout_name: Label Test
     guidebox: 0
oblique_title: 1
         font: TR
    font_size: 3
  scale_width: 0.080000
 scale_height: 0.010000
        units: POINT
callnum_split: 0
 text_justify: L
format_string: title, author, isbn, issn, itemtype, barcode, itemcallnumber
   layout_xml: <opt></opt>
      creator: Labels
1 row in set (0.001 sec)

Layout 17 after step 14

MariaDB [koha_kohadev]> select * from creator_layouts where layout_id = 17 \G;
*************************** 1. row ***************************
    layout_id: 17
 barcode_type: CODE39
  start_label: 1
printing_type: BIBBAR
  layout_name: Label Test
     guidebox: 0
oblique_title: 1
         font: TR
    font_size: 3
  scale_width: 0.800000
 scale_height: 0.010000
        units: POINT
callnum_split: 0
 text_justify: L
format_string: title, author, isbn, issn, itemtype, barcode, itemcallnumber
   layout_xml: <opt></opt>
      creator: Labels
1 row in set (0.000 sec)

Oh I see it *facepalm*

Before : 0.080000
After  : 0.800000

Created attachment 166443 [details] [review]
Bug 36819: Default layout data prints squished barcodes

This patch changes the default layout data for the barcode
width.

To test:
**On a fresh db**
1. Go to Cataloging > Label creator
2. Click New > Label batch
3. Enter the following barcodes in the text field (or enter your own)

39999000010114
39999000010138
39999000010152
39999000011333
39999000002331
39999000002355

4. Click Add items
5. Click Export full batch
6. Choose the following
   - Template: Avery 5160 | 1 x 2-5/8
   - Layout: Label test
7. Click Export
8. Click Download as PDF
9. Open the PDF
   --> Note the squished barcodes
10. Apply patch and reset_all
11. Redo steps 1 to 9
    --> Note the barcodes now look OK

Created attachment 166667 [details] [review]
Bug 36819: Default layout data prints squished barcodes

This patch changes the default layout data for the barcode
width.

To test:
**On a fresh db**
1. Go to Cataloging > Label creator
2. Click New > Label batch
3. Enter the following barcodes in the text field (or enter your own)

39999000010114
39999000010138
39999000010152
39999000011333
39999000002331
39999000002355

4. Click Add items
5. Click Export full batch
6. Choose the following
   - Template: Avery 5160 | 1 x 2-5/8
   - Layout: Label test
7. Click Export
8. Click Download as PDF
9. Open the PDF
   --> Note the squished barcodes
10. Apply patch and reset_all
11. Redo steps 1 to 9
    --> Note the barcodes now look OK

Signed-off-by: David Nind <david@davidnind.com>

Created attachment 166734 [details] [review]
Bug 36819: Default layout data prints squished barcodes

This patch changes the default layout data for the barcode
width.

To test:
**On a fresh db**
1. Go to Cataloging > Label creator
2. Click New > Label batch
3. Enter the following barcodes in the text field (or enter your own)

39999000010114
39999000010138
39999000010152
39999000011333
39999000002331
39999000002355

4. Click Add items
5. Click Export full batch
6. Choose the following
   - Template: Avery 5160 | 1 x 2-5/8
   - Layout: Label test
7. Click Export
8. Click Download as PDF
9. Open the PDF
   --> Note the squished barcodes
10. Apply patch and reset_all
11. Redo steps 1 to 9
    --> Note the barcodes now look OK

Signed-off-by: David Nind <david@davidnind.com>

Created attachment 166735 [details] [review]
Bug 36819: Default layout data prints squished barcodes

This patch changes the default layout data for the barcode
width.

To test:
**On a fresh db**
1. Go to Cataloging > Label creator
2. Click New > Label batch
3. Enter the following barcodes in the text field (or enter your own)

39999000010114
39999000010138
39999000010152
39999000011333
39999000002331
39999000002355

4. Click Add items
5. Click Export full batch
6. Choose the following
   - Template: Avery 5160 | 1 x 2-5/8
   - Layout: Label test
7. Click Export
8. Click Download as PDF
9. Open the PDF
   --> Note the squished barcodes
10. Apply patch and reset_all
11. Redo steps 1 to 9
    --> Note the barcodes now look OK

Signed-off-by: David Nind <david@davidnind.com>
Signed-off-by: Lucas Gass <lucas@bywatersolutions.com>

Did we figure out why this change became necessary? Should we have a note for existing installations or even a database update?

Pushed for 24.05!

Well done everyone, thank you!

(In reply to Katrin Fischer from comment #7)
> Did we figure out why this change became necessary? Should we have a note
> for existing installations or even a database update?

The existing updatedb was correct. It was only for new installations that there was a problem. Is there a way to discriminate the ones that are ok and the ones that were installed since the bug.

Thx Caroline. I think maybe we can do without an additional update then as we can not really be sure if the value is 'original' or has been changed?

(In reply to Katrin Fischer from comment #10)
> Thx Caroline. I think maybe we can do without an additional update then as
> we can not really be sure if the value is 'original' or has been changed?

Can we do "IF value = 0.080000 THEN change value to 0.800000" ?

We can do something like:

UPDATE creator_layouts SET scale_width = 0.800000 WHERE scale_width = 0.080000;

We cannot really tell if it's an unchanged sample one... but maybe that's a very low risk here.

(In reply to Katrin Fischer from comment #12)
> We can do something like:
> 
> UPDATE creator_layouts SET scale_width = 0.800000 WHERE scale_width =
> 0.080000;
> 
> We cannot really tell if it's an unchanged sample one... but maybe that's a
> very low risk here.

Just catching up here.

This really should not have been pushed without the constraint Katrin proposes. the time we think it won't matter, it will. That's just how things roll unfortunately.

We haven't released with the first patch yet, so there is still time for the database update, but the problem started earlier with bug 10762.

Also: Hi Chris N.! *waves*

(In reply to Katrin Fischer from comment #15)
> Also: Hi Chris N.! *waves*

Hey Katrin! *waves back*

Created attachment 166974 [details] [review]
Bug 36819: (follow-up) Updatedatabase to correct wrong value

This patch adds an update to correct the value of scale_width to
0.800000 IF it is 0.080000. It should not change the value if it
is anything else than 0.080000.

Created attachment 166982 [details] [review]
Bug 36819: (follow-up) Updatedatabase to correct wrong value

This patch adds an update to correct the value of scale_width to
0.800000 IF it is 0.080000. It should not change the value if it
is anything else than 0.080000.

Signed-off-by: Martin Renvoize <martin.renvoize@ptfs-europe.com>

Created attachment 166983 [details] [review]
Bug 36819: (QA follow-up) Improve update output

Hm, seeing this, I'll try to see if I can understand/fix:

 FAIL	installer/data/mysql/atomicupdate/bug_36819.pl
   FAIL	  valid
		syntax error
		installer/data/mysql/atomicupdate/bug_36819.pl had compilation errors.

Pushed for 24.05!

Well done everyone, thank you!

(In reply to Katrin Fischer from comment #20)
> Hm, seeing this, I'll try to see if I can understand/fix:
> 
>  FAIL	installer/data/mysql/atomicupdate/bug_36819.pl
>    FAIL	  valid
> 		syntax error
> 		installer/data/mysql/atomicupdate/bug_36819.pl had compilation errors.

Did you figure it out / fix it?

Yes, I did - it was a stray semicolon in the follow-up. Fixed and pushed.

For release maintainers, this came about in 23.11. Before that, the scale_width doesn't exist.

Looks like 'additional_work_needed' is obsolete

I prefer looks at this in next release 23.11.07

(In reply to Fridolin Somers from comment #25)
> Looks like 'additional_work_needed' is obsolete

I think so too. I don't see what else is missing. I removed the keyword.

Pushed to 23.11.x for 23.11.07

I've added a release note, even though it was pushed to 24.05.

I've also changed the status to "Needs documenting", as this is a fix for an enhancement in 23.11.00 that was not back ported to any earlier releases.

Not backporting to 23.05.x unless requested

This is a bug fix, no documentation needed.