They work the same way as conditional filenames (84d2e640), e.g.
"directory": {
"score >= 20": ["high score"],
"score >= 5" : ["mid score"],
"" : ["{category}", "default"]
}
to convert a timestamp to a formattable 'datetime' object.
For example '{created_at!d:%Y-%m-%d}'
transforms the timestamp in 'created_at' into a 'datetime' object
and then formats its content using '%Y-%m-%d' as template.
1262304000 -> datetime(2010, 1, 1) -> "2010-01-01"
Separating the archive check from pathfmt.exists() in b5243297
had some unintended side effects.
It is also not possible to monkey-patch a dunder method like
__contains__ because of the special method lookup that gets
performed for them.
… and not just None (#525)
It would be better to consistently use None for all non-existent
fields and/or fields without a valid value, but this is a good
enough workaround for now.
- fix infinite recursion for responses with multiple entries in
'history'
- hide values of Set-Cookie headers
- only write the response content by default
(use '-o write-pages=all' to also include HTTP headers)
Allow its value to be a JSON object / Python dict that specifies
a mapping from invalid/unwanted input characters to specific
output characters.
For example {"/": "-", "*": "+"} will transform
"foo / ***bar***" into "foo - +++bar+++"
(closes#662, #755)
Wrap all loggers used by job, extractor, downloader, and postprocessor
objects into a (custom) LoggerAdapter that provides access to the
underlying job, extractor, pathfmt, and kwdict objects and their
properties.
__init__() signatures for all downloader and postprocessor classes have
been changed to take the current Job object as their first argument,
instead of the current extractor or pathfmt.
(#574, #575)
The format string '{a|b|c}' will now try to use the value from 'a' and
fall back to 'b' and 'c' if accessing a field raises an exception or
if its value is None.
- take a file object as argument instead of an filename
- accept whitespace before comments (" # comment")
- map expiration "0" to None and not the number 0
The methods of the standard libraries' MozillaCookieJar have
several shortcomings (#HttpOnly_ cookies, 0 expiration timestamps, etc.)
and require construction of an ultimately pointless CookieJar object.
- use str.join() instead of os.path.join()
(less "features", but 10x as fast)
- cache directory formatters
- detect and optimize field access for 1-element format strings
- change 'has_extension' from a simple flag/bool to a field that
contains the original filename extension
- rename 'keywords' to 'kwdict' and some other stuff as well
- inline 'adjust_path()'
- put enumeration index before filename extension (#306)
Simplified universal serialization support in json.dump() can be achieved
by passing 'default=str', which was already the case in DataJob.run()
for -j/--dump-json, but not for the 'metadata' post-processor.
This commit introduces util.dump_json() that (more or less) unifies the
JSON output procedure of both --write-metadata and --dump-json.
(#251, #252)
This allows for stuff like "{extractor.url}" and "{extractor.category}"
in logging format strings.
Accessing 'extractor' and 'job' in any way will return "None" if those
fields aren't defined, i.e. in general logging messages.
In addition to 'abort' and 'exit', it is now possible to specify
'abort:N' and 'exit:N' (where N is any integer) as value for 'skip'
to abort/exit after consecutively skipping N downloads.
The functionality of --(chapter-)filter and --(chapter-)range are now
also exposed as the following config-file options:
- extractor.*.image-filter
- extractor.*.image-range
- extractor.*.chapter-filter
- extractor.*.chapter-range
TODO: update configuration.rst
A format string now gets parsed only once instead of re-parsing it each
time it is applied to a set of data.
The initial parsing causes directory path creation to be at about 2x
slower than before, since each format string there is used only once,
but building a filename, the more common operation, is at least 2x
faster. The "directory slowness" cancels at about 5 filenames and
everything above that is significantly faster.
The L option allows for the contents of a format field to be replaced
with <replacement> if its length is greater than <maxlen>.
Example:
{f:L5/too long/} -> "foo" (if "f" is "foo")
-> "too long" (if "f" is "foobar")
(#92) (#94)
It is now possible to slice string (or list) values of format string
replacement fields with the same syntax as in regular Python code.
"{digits}" -> "0123456789"
"{digits[2:-2]}" -> "234567"
"{digits[:5]}" -> "01234"
The optional third parameter (step) has been left out to simplify things.
OAuth support for SmugMug needs some additional features
(auth-rebuild on redirect, query parameters in URL, ...)
and fixing this in the old code wouldn't work all that well.
Mike Fährmann