Mythfilldatabase translates all program genres to lower case

[John Bergqvist <JohnLBergqvist@…>]

When using the atlas or radio times XMLTV grabbers, mythfilldatabase is translating all genres/categories (and possibly category_type too) for each program into lowercase when they are inserted into the program table.

When running the grabbers manually & inspecting the XML files, the genres are in their correct case & capitalised as appropriate, but when mythfilldatabase inserts them into the database, any capitalisation is lost. This does not occur for genres of programs inserted via EIT, so it looks to me that this is mythfilldatabase's doing.

[JohnLBergqvist@…]

Proposed pull request: ​https://github.com/MythTV/mythtv/pull/108

[William L. DeRieux IV <williamderieux@…>]

Why not just use a case-insensitive comparison?

if ((cat.compare(QObject::tr("movie"),Qt::CaseSensitivity::CaseInsensitive) == 0) ||
        (cat.compare(QObject::tr("film"),Qt::CaseSensitivity::CaseInsensitive) == 0))
{
    // Hack for tv_grab_uk_rt
    pginfo->categoryType = ProgramInfo::kCategoryMovie;

[William L. DeRieux IV <williamderieux@…>]

Replying to William L. DeRieux IV <williamderieux@…>:

Why not just use a case-insensitive comparison?

if ((cat.compare(QObject::tr("movie"),Qt::CaseSensitivity::CaseInsensitive) == 0) ||
        (cat.compare(QObject::tr("film"),Qt::CaseSensitivity::CaseInsensitive) == 0))
{
    // Hack for tv_grab_uk_rt
    pginfo->categoryType = ProgramInfo::kCategoryMovie;

updated wrong use of Qt::CaseSensitivity::CaseInsensitive

if ((cat.compare(QObject::tr("movie"),Qt::CaseInsensitive) == 0) ||
        (cat.compare(QObject::tr("film"),Qt::CaseInsensitive) == 0))
{
    // Hack for tv_grab_uk_rt
    pginfo->categoryType = ProgramInfo::kCategoryMovie;

[JohnLBergqvist@…]

My change was solely built on the existing code.

[William L. DeRieux IV <williamderieux@…>]

Replying to JohnLBergqvist@…:

My change was solely built on the existing code.

True I realize that. But a case-insensitive comparison will future proof the code should the category case be changed.

[Gary Buhrmaster <gary.buhrmaster@…>]

Note for any potential committer: This pull request appears to add/change translated strings (which may have an impact on any proposed back-porting and the translators workload and processes (*)).

(*) I admit to knowing very little about the translators work flow, but I do know there is work there.

[Nicolas Riendeau]

Hi Gary!

Changing string at this point in time is still acceptable if it fixes a problem but will no longer be in a few weeks time when we go into full string freeze (we are in soft string freeze right now).

Don't expect any updates from our translators against 0.27 though...

Since it doesn't change the strings *and* as William said it's more future proof I personally much rather prefer the case-insensitive comparison...

Thank you and have a nice day!

Nicolas Riendeau MythTV developer and translation maintainer

[Nicolas Riendeau]

I was contacted about this so I better explain what I said above a little better...

Assuming it does what it is supposed to, I am definitely AGAINST backporting it to 0.27 as it would break translations, while it might fix one person's problem it would create one for someone else...

Our translators currently work against master and the upcoming 0.28, we very rarely receive updates for the current -fixes version so it would create a problem which would not get fixes until the next -fixes version.

However, if the translatable strings are left as they initially were and the comparison is made case-insensitive and assuming this does what it is supposed to I have no problem with it being backported to 0.27.

I also think that since we have no control over the data we are feeded we should make the comparison case-insensitive and not depend on it being of a specific case, this could likely break in the future...

John, could you post a revised patch?

Thank you and have a nice day!

Nicolas

[William L. DeRieux IV <williamderieux@…>]

Replying to knight:

I was contacted about this so I better explain what I said above a little better...

Assuming it does what it is supposed to, I am definitely AGAINST backporting it to 0.27 as it would break translations, while it might fix one person's problem it would create one for someone else...

Our translators currently work against master and the upcoming 0.28, we very rarely receive updates for the current -fixes version so it would create a problem which would not get fixes until the next -fixes version.

However, if the translatable strings are left as they initially were and the comparison is made case-insensitive and assuming this does what it is supposed to I have no problem with it being backported to 0.27.

I also think that since we have no control over the data we are feeded we should make the comparison case-insensitive and not depend on it being of a specific case, this could likely break in the future...

John, could you post a revised patch?

Thank you and have a nice day!

Nicolas

maybe this should marked as 'info needed' since an updated patch has been requested.

[JohnLBergqvist@…]

Updated patch posted into pull request: ​https://github.com/MythTV/mythtv/pull/108

[John Bergqvist <JohnLBergqvist@…>]

Replying to knight:

I was contacted about this so I better explain what I said above a little better...

Assuming it does what it is supposed to, I am definitely AGAINST backporting it to 0.27 as it would break translations, while it might fix one person's problem it would create one for someone else...

Our translators currently work against master and the upcoming 0.28, we very rarely receive updates for the current -fixes version so it would create a problem which would not get fixes until the next -fixes version.

However, if the translatable strings are left as they initially were and the comparison is made case-insensitive and assuming this does what it is supposed to I have no problem with it being backported to 0.27.

I also think that since we have no control over the data we are feeded we should make the comparison case-insensitive and not depend on it being of a specific case, this could likely break in the future...

John, could you post a revised patch?

Thank you and have a nice day!

Nicolas

I've posted an updated patch into my pull request: ​​https://github.com/MythTV/mythtv/pull/108

[JohnBergqvist <John@…>]

In c8eecc119925fc26749750619db23623b335615f/mythtv:

Error: Processor CommitTicketReference failed

GIT backend not available

[Nicolas Riendeau <nriendeau@…>]

In 702d0238cf7d8e4c7d6d4d749cc098a02a6616d1/mythtv:

Error: Processor CommitTicketReference failed

GIT backend not available

[Nicolas Riendeau <nriendeau@…>]

In a655a8762f53944be8588db9b9321e0be5edebdf/mythtv:

Error: Processor CommitTicketReference failed

GIT backend not available

[Nicolas Riendeau <nriendeau@…>]

In d76c3518c23f2e8abbdb7debac56b521e3a2f53e/mythtv:

Error: Processor CommitTicketReference failed

GIT backend not available

[Nicolas Riendeau]

Thank you John!