Templates: various questions not worth their own thread

[ownedbycats]

In my specific case, I've started copying the previous last-read dates to #admintags, as 'read:2020-04-16'.

Here's the template I used to make the list:

program:

	most_recent= format_date($$#lastread, 'yyyy-MM-dd');
	old_dates= list_re($#admintags, ',', 'read:', ' ');

	date_list= list_sort(list_union(most_recent, old_dates, ','), 0, ',')

Unfortunately, short of editing the read: tag entries I'm not sure if it's possible to do it while building the list.

Additionally: is using list_re to replace 'read:' with a space the proper method?

[chaley]

Originally Posted by ownedbycats

Unfortunately, short of editing the read: tag entries I'm not sure if it's possible to do it while building the list.

It is certainly possible, but without knowing the processing sequence(s) and seeing all the processing I can't know what would be the most efficient. And it might not matter if these are action chains templates you run on-demand.

Additionally: is using list_re to replace 'read:' with a space the proper method?

In this template, probably. If 'read:' can appear only has the beginning of a date tag then

old_dates= re($#admintags, ',', 'read:', ' ');

works and would be faster.

[ownedbycats]

Due to people posting non-template questions here, I made a new thread for small library management questions:

https://www.mobileread.com/forums/sh...d.php?t=355273

[theodduck]

What would be the best way to remove/replace a list of special characters from titles when saving to disk/sending to device? I'd like to make sure certain characters like colons and commas aren't included and I don't like that Calibre defaults to replacing colons with an underscore. I'm not a coder so I'm kind of fumbling my way through this.

So far I've stumbled through some posts and the documentation to get the following to replace colons with nothing, but I'm not sure how to make it do the same for additional characters

{title: re(":","")}

By some miracle, I've come up with the following to do the above AND limit the file name length, so if the solution could be incorporated into this, that'd be even better.

{title:'sublist(re($, ":", ""), 0, 8, " ")'}

If there is a different way I should be doing this, let me know.

[capink]

Originally Posted by theodduck

What would be the best way to remove/replace a list of special characters from titles when saving to disk/sending to device? I'd like to make sure certain characters like colons and commas aren't included and I don't like that Calibre defaults to replacing colons with an underscore. I'm not a coder so I'm kind of fumbling my way through this.

So far I've stumbled through some posts and the documentation to get the following to replace colons with nothing, but I'm not sure how to make it do the same for additional characters

{title: re(":","")}

{title: re("[:\,]","")}

The comma here is escaped with a backslash. Some characters don't need to be escaped. For example if you want to also remove characters like semicolons and underscores.

{title: re("[:\,;_]","")}

Trial and error should guide you into what needs to be escaped and what does not need.

[theodduck]

Originally Posted by capink

(snip)

Yes, this works great! Thanks for pointing me in the right direction.

[ownedbycats]

Is there any functional difference between $$#timesread>#0 or $$#timesread>=#1?

program:
	status = readstatus();
	times = $$#timesread;

	switch_if(
		status=='currentlyreading' && times>#0, 'Currently Rereading',
		status=='currentlyreading', 'Currently Reading',
		status=='toberead' && times>#0, 'To Be Reread',
		status=='toberead', 'To Be Read',
		status=='read', 'Read',
		status=='unread', 'Unread',
		status=='didnotfinish', 'Did Not Finish',
		''
	)

[chaley]

Originally Posted by ownedbycats

Is there any functional difference between $$#timesread>#0 or $$#timesread>=#1?

program:
	status = readstatus();
	times = $$#timesread;

	switch_if(
		status=='currentlyreading' && times>#0, 'Currently Rereading',
		status=='currentlyreading', 'Currently Reading',
		status=='toberead' && times>#0, 'To Be Reread',
		status=='toberead', 'To Be Read',
		status=='read', 'Read',
		status=='unread', 'Unread',
		status=='didnotfinish', 'Did Not Finish',
		''
	)

No difference in meaning or performance. There are usually multiple ways to express a relational conditional. Often one of them makes more sense to a reader. In your case, to me ">#0" better expresses the logic.

You could even do

! $$#timesread<=#0

[ownedbycats]

Just out of curiosity, I put this in the template tester:

program:

mydate = '2000';

format_date(mydate, 'yyyy-MM-dd');

I always end up with [year]-10-15. Where does the October 15th date come from? I thought it'd be the current date, but it's only the 11th where I live.

[chaley]

Originally Posted by ownedbycats

Just out of curiosity, I put this in the template tester:



I always end up with [year]-10-15. Where does the October 15th date come from? I thought it'd be the current date, but it's only the 11th where I live.

The template processor uses calibre's date parser, which is dateutil. When it parses a date, missing fields come from "default". In calibre, default is specified as

func = datetime.utcnow if assume_utc else datetime.now
default = func().replace(day=15, hour=0, minute=0, second=0, microsecond=0,
                tzinfo=_utc_tz if assume_utc else _local_tz)

The variable assume_utc is False.

As such, a missing month is the current local month and a missing day is the 15th. A missing time becomes midnight. I don't know if one can have a missing year, but if so then it would be the current year.

[ownedbycats]

program:

dates = list_count($#readdates, ',');

readcount = $$#timesread;

if dates != readcount then 'yes' fi

When both #readdates (tag-like) and timesread (integer) are undefined, this returns 'yes.' Why?

[chaley]

Originally Posted by ownedbycats

program:

dates = list_count($#readdates, ',');

readcount = $$#timesread;

if dates != readcount then 'yes' fi

When both #readdates (tag-like) and timesread (integer) are undefined, this returns 'yes.' Why?

Because the string '0' (zero, list_count()) doesn't equal the string 'none' (raw_field()). If you used !=# then IIRC zero would equal none.

[ownedbycats]

Ah yeah, adding a hash sign fixes that. Thanks.

[capink]

When writing a template function, is there a way to test whether the function is called from a python template or not? If so, can it return a python object (e.g. python list)? or would it result in an error?

[chaley]

Originally Posted by capink

When writing a template function, is there a way to test whether the function is called from a python template or not?

Not easily. You can examine the stack and see if "call" is method name in the calling frame, but I think this might be brittle. See below for another idea.

If so, can it return a python object (e.g. python list)? or would it result in an error?

Yes, a python template function can return a python object. The formatter returns it to the calling template unchanged.

I suggest you provide two functions, one that returns a string and one that returns whatever. Alternately, provide an optional parameter that tells the function what to return.

Note that if a python template function returns a list then the formatter tries to convert the returned value to a string with ', '.join(returned_value). Of course this only works if the list contains strings.