Fix saving of multiple TempoSyncKnobModels by zonkmachine · Pull Request #3281 · LMMS/lmms

zonkmachine · 2017-01-21T21:49:16Z

zonkmachine · 2017-01-21T21:50:26Z



+
+void DataFile::upgrade_1_2_0_rc2()


What should this be? DataFile::upgrade_1_2_0_rc3() ?

zonkmachine · 2017-01-21T23:29:18Z

Fixed. I put the upgrade path in TempoSyncKnobModel::loadSettings() as it would have grown in complexity otherwise.

zonkmachine · 2017-01-24T01:54:35Z

I'll merge this in a day or so if there are no objections.

jasp00 · 2017-01-24T09:34:41Z

I put the upgrade path in TempoSyncKnobModel::loadSettings() as it would have grown in complexity otherwise.

What do you mean? loadSettings() has grown in complexity. We are supposed to move away from these patterns. Upgrades belong to DataFile.

zonkmachine · 2017-01-25T14:45:06Z

What do you mean? ...

I was a bit tired and got lost in translation. I meant that using DataFile.cpp would be, in this case, much more difficult.
How do you pair syncmode with it's knob? You need to iterate through the list elements and if you find syncmode you know that at least one of the elements is a sync knob but which?

Monstro save file, lmms-1.1.3

monstro o1pw="50" v1e2="0" e2sus="1" l1att_numerator="4" e2rel="0" l1wav="0" e2dec_denominator="4"
s3e1="0" o3vol="33" s3e2="0" w1l1="0" o2spo="0" w1l2="0" o2crs="0" o1ssf="0" o1pan="0" p3l1="0" o2wav="0"
p3l2="0" l2rat_denominator="4" o2ftl="0" e1dec_numerator="4" e2hol_denominator="4" p2l1="0" p2l2="0"
o3wav1="0" o2syn="0" l1phs="0" o3wav2="0" o2ftr="0" o1ssr="0" p1l1="0" p1l2="0" w1e1="0" e2hol_numerator="4"
w1e2="0" l1rat="1" l1rat_numerator="4" o2synr="0" p3e1="0" p3e2="0" e1att_numerator="4"
e1dec_denominator="4" l2wav="0" p2e1="0" p2e2="0" e1dec="0" e1rel_denominator="4" l1att="1714" o3spo="0"
p1e1="0" o3crs="0" p1e2="0" e2pre_numerator="4" e1pre="0" o2pan="0" e1hol_denominator="4"
syncmode="0" l2att_denominator="4" e1att="0" e1hol="0" o3syn="0" f3l1="0" l2att_numerator="4" l2phs="0"
f3l2="0" e2rel_numerator="4" f2l1="0" f2l2="0" e1pre_denominator="4" l2rat="1" e2att_denominator="4"
o3synr="0" e2rel_denominator="4" o1vol="33" f1l1="0" f1l2="0" e1slo="0" e2dec_numerator="4" e2dec="0"
f3e1="0" f3e2="0" l2att="429" f2e1="0" e2pre="0" f2e2="0" o3pan="0" o23mo="0" l1att_denominator="4" f1e1="0"
e2att="0" f1e2="0" e2hol="0" e1sus="1" e2pre_denominator="4" v3l1="0" v3l2="0" e2att_numerator="4" e1rel="0"
e1att_denominator="4" v2l1="0" v2l2="0" l2rat_numerator="4" e1hol_numerator="4" o2vol="33" v1l1="0" o1spo="0"
v1l2="0" o1crs="0" e1pre_numerator="4" e2slo="0" o3sub="0" v3e1="0" v3e2="0" o1ftl="0" v2e1="0" v2e2="0"
s3l1="0" o1ftr="0" s3l2="0" l1rat_denominator="4" e1rel_numerator="4" v1e1="0"

Monstro save file, lmms-1.2.0-temposyncknobmodel branch

monstro e1slo="0" e2pre_numerator="4" e2dec_syncmode="3" s3e1="0" s3e2="0" o3vol="33"
e1rel_numerator="4" l1rat_syncmode="3" o2vol="33" l2att_syncmode="3" e2hol="857.143" o3spo="0"
e2hol_syncmode="3" l1att_numerator="4" o1vol="33" e1hol="857.143" o2spo="0" o3pan="0" p3e1="0" p3e2="0"
p2e1="0" w1l1="0" p2e2="0" p1e1="0" p1e2="0" w1l2="0" o1pw="50" e1dec_numerator="4" e2pre="857.143"
o23mo="0" o1spo="0" l2wav="0" o2pan="0" o2wav="0" l2rat="857.143" e1att_numerator="4" f3l1="0" f3l2="0"
f2l1="0" v3l1="0" f2l2="0" f1l1="0" v3l2="0" v2l1="0" f1l2="0" v2l2="0" v1l1="0" v1l2="0" l1att_denominator="4"
e1dec_denominator="4" e1pre="857.143" l1rat_denominator="4" l1wav="0" e1att_denominator="4" o1pan="0"
l1rat="857.143" o3crs="0" e2rel_numerator="4" e1pre_syncmode="3" e1pre_denominator="4"
e2rel_syncmode="3" o3sub="0" o2crs="0" o1ssf="0" l2att_numerator="4" o1ssr="0" o1crs="0" e2rel="857.143"
e1rel_denominator="4" s3l1="0" s3l2="0" e2sus="1" o2ftl="0" e2dec="857.143" e2att="857.143"
e2dec_numerator="4" o2ftr="0" l2att="857.143" e2att_numerator="4" e1rel="857.143" l1rat_numerator="4"
e1dec="857.143" e1sus="1" o1ftl="0" e1hol_numerator="4" e1att="857.143" l2att_denominator="4" o1ftr="0"
l2rat_denominator="4" e2dec_denominator="4" l1att="857.143" e2att_denominator="4" e1hol_denominator="4"
e1att_syncmode="3" e2pre_denominator="4" p3l1="0" l2rat_syncmode="3" p3l2="0" p2l1="0" p2l2="0"
p1l1="0" p1l2="0" o2synr="0" o3syn="0" o3synr="0" l2phs="0" e2rel_denominator="4" e1dec_syncmode="3"
e1pre_numerator="4" o2syn="0" l1phs="0" o3wav1="0" o3wav2="0" l2rat_numerator="4" e2hol_numerator="4"
w1e1="0" w1e2="0" e1rel_syncmode="3" f3e1="0" f3e2="0" f2e1="0" v3e1="0" l1att_syncmode="3" f2e2="0"
f1e1="0" v3e2="0" v2e1="0" f1e2="0" v2e2="0" v1e1="0" v1e2="0" e2hol_denominator="4" e2att_syncmode="3"
e1hol_syncmode="3" e2slo="0" e2pre_syncmode="3"

jasp00 · 2017-01-26T08:38:29Z

if you find syncmode you know that at least one of the elements is a sync knob but which?

All the ones that have a meter specification. The upgrade step would add a name_syncmode for every name_numerator.

zonkmachine · 2017-01-28T08:20:18Z

Thanks!

Getting there slowly... 🐌
I'm pushing a half finished upgrade function for inspection.
I need to find elements that contain "syncmode".
If I've got this anywhere near correct I think the way to proceed from here now it to modify the function void findIds(const QDomElement& elem, QList<jo_id_t>& idList) in DataFile.cpp
to take a string argument and make it generally hand back lists of elements when the node containing that element. It's now hardwired to "id".
https://github.com/LMMS/lmms/blob/master/src/core/DataFile.cpp#L1088:L1100

zonkmachine · 2017-01-28T08:27:55Z

+	QDomNode node;
+
+	// syncmode for multiple tempo sync knobs
+	list = elementsByTagName( "syncmode" );


So, this isn't working. The function I'm looking for would look something like:
list = elementThatHasAttribute( "syncmode" );

zonkmachine · 2017-01-28T08:29:00Z

 	void upgrade_1_0_99();
 	void upgrade_1_1_0();
 	void upgrade_1_1_91();
+	void upgrade_1_2_0_rc2_23();


What's a better name for the next upgrade function?

zonkmachine · 2017-01-28T08:36:21Z

+				{
+					name = name.remove( "_numerator" );
+					name = name.append( "_syncmode" );
+					el.setAttribute( name, syncmode );


This should rather set only the last attribute "syncmode" in every element as only the last one get's saved currently. The other attributes wouldn't get written over as they are today.

zonkmachine · 2017-02-03T12:01:34Z

@LMMS/developers
Unfortunately I'm not going to be able to solve this within a reasonable time limit. I'm in fact well beyond that limit at this point. Not without a level of support that makes it better for someone else to do it.

I revert to my original working commit and leave the best attempts at upgrade paths (incomplete) as gists. I think probably attempt 2 is the way to go...
temposyncbug upgrade attempt 1
temposyncbug upgrade attempt 2

jasp00 · 2017-02-04T20:48:14Z

Upgrade methods should be run until the build version they were introduced. Pull requests can only estimate this version. A maintenance task should take care of this problem; I know how it should work.

zonkmachine

I've tested this and it works very well.

tresf · 2017-02-05T03:29:00Z

Upgrade methods should be run until the build version they were introduced. Pull requests can only estimate this version. A maintenance task should take care of this problem; I know how it should work.

Please no.

jasp00 · 2017-02-05T06:01:16Z

Please no.

Why not? We can easily allow upgrades from development versions, without implying official support.
(Do I open a separate issue?)

tresf · 2017-02-05T14:52:34Z

It's overkill. We only need to upgrade with RCs at the most granular of levels. We don't release any more granular than RCs, it's plenty.

jasp00 · 2017-02-13T17:13:49Z

This still works. I will merge in some hours.

* Save multiple TempSyncKnobModel syncmodes * Add upgrade path for approximate version 1.2.0-rc2.42

zonkmachine added the in progress label Jan 21, 2017

zonkmachine commented Jan 21, 2017

View reviewed changes

zonkmachine force-pushed the temposyncknobmodel branch from f5758c3 to 276fc35 Compare January 21, 2017 23:23

zonkmachine removed the in progress label Jan 21, 2017

zonkmachine force-pushed the temposyncknobmodel branch from 276fc35 to d386d4b Compare January 28, 2017 08:11

zonkmachine commented Jan 28, 2017

View reviewed changes

zonkmachine force-pushed the temposyncknobmodel branch from d386d4b to 873d6c4 Compare February 3, 2017 11:59

jasp00 mentioned this pull request Feb 4, 2017

Move private functions to private files (example) #3213

Closed

zonkmachine commented Feb 4, 2017

View reviewed changes

zonkmachine mentioned this pull request Feb 8, 2017

Beat note bug at the end of bars #2964

Merged

jasp00 mentioned this pull request Feb 13, 2017

Allow continuous use of master version #3347

Closed

zonkmachine and others added 2 commits February 13, 2017 03:04

Save multiple TempSyncKnobModel syncmodes

e120555

Add upgrade path for approximate version 1.2.0-rc2.42

bce2b5e

jasp00 force-pushed the temposyncknobmodel branch from 94e077c to bce2b5e Compare February 13, 2017 02:14

jasp00 merged commit c1321ba into LMMS:master Feb 14, 2017

zonkmachine deleted the temposyncknobmodel branch March 13, 2017 18:18

sdasda7777 pushed a commit to sdasda7777/lmms that referenced this pull request Jun 28, 2022

Fix saving of multiple TempoSyncKnobModels (LMMS#3281)

1c5e087

* Save multiple TempSyncKnobModel syncmodes * Add upgrade path for approximate version 1.2.0-rc2.42

Uh oh!

Uh oh!

Conversation

zonkmachine commented Jan 21, 2017

Uh oh!

zonkmachine Jan 21, 2017

Choose a reason for hiding this comment

Uh oh!

zonkmachine commented Jan 21, 2017

Uh oh!

zonkmachine commented Jan 24, 2017

Uh oh!

jasp00 commented Jan 24, 2017

Uh oh!

zonkmachine commented Jan 25, 2017

Uh oh!

jasp00 commented Jan 26, 2017

Uh oh!

zonkmachine commented Jan 28, 2017

Uh oh!

zonkmachine Jan 28, 2017

Choose a reason for hiding this comment

Uh oh!

zonkmachine Jan 28, 2017

Choose a reason for hiding this comment

Uh oh!

zonkmachine Jan 28, 2017

Choose a reason for hiding this comment

Uh oh!

zonkmachine commented Feb 3, 2017 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

jasp00 commented Feb 4, 2017

Uh oh!

zonkmachine left a comment

Choose a reason for hiding this comment

Uh oh!

tresf commented Feb 5, 2017

Uh oh!

jasp00 commented Feb 5, 2017

Uh oh!

tresf commented Feb 5, 2017

Uh oh!

jasp00 commented Feb 13, 2017

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

zonkmachine commented Feb 3, 2017 •

edited

Loading