pictuga
84377a3f36
Commit "id" in feedify
...
Has been there for a while, never bothered to commit.
9 years ago
pictuga
3e2c4de833
Fix bug in parseOptions with values
9 years ago
pictuga
bd1c6a5be6
Huge commit: wsgiref-like. "Progress" dropped.
...
Can now run as HTTP server. Main functions split again (Init, Fetch). LIM/MAX_STUFF moved to function body (this still has to be improved). "progress" too hard to get to work with wsgiref, dropped. This whole thing should make it easier to maintain code (no more ifs everywhere).
9 years ago
pictuga
f12031cbdb
More usefull except logging
9 years ago
pictuga
a2402da32e
Stop logging accept-header string
...
Useless once stable
9 years ago
pictuga
69a8516cd1
Slightly nicer "rank" computer in accept-header
9 years ago
pictuga
32f9550be5
Add "al" (apparent limit) param with value
...
Removes element from rss *after* treatment, useful for slow json API
9 years ago
pictuga
e70399b654
Raise MorssExceptions in morss.py
...
Cleaner
9 years ago
pictuga
47b8c12fbb
Remove useless 'socket' dependence
9 years ago
pictuga
bb48ee4518
Forgot one chunk of code in previous commit
9 years ago
pictuga
a5e4775f68
Move some code into Gather()
...
Will be useful when playing with WSGIREF
9 years ago
pictuga
b86b0f371a
Remove silly debug print
9 years ago
pictuga
0badec72a9
Allow iteration over RSS objects
...
Useful for JSON
9 years ago
pictuga
d97a72bc0b
Use Cache dict for facebook storage
9 years ago
pictuga
afe31e695a
"json" arg to output json feed
9 years ago
pictuga
47013c7948
Use sys.stdout.write instead of print in progress
...
For some reason, this avoids having multiple "lines" on top of each other, which was always due to print()' newlines. And this makes the client/js part much more stable.
9 years ago
pictuga
eafbc14a58
Allow "progress" in JSON
9 years ago
pictuga
cbdb61a918
Allow real python dict in Cache
...
Makes it possible to store things a nicer way
9 years ago
pictuga
2a8c94fd88
Work with unicode strings
...
In downloads at least (I don't quite remember what readability returns actually, but "it just works")
9 years ago
pictuga
eac2e7a79a
Try to avoid erros with saving
...
Mainly: do not create empty cache file, and accept empty cache file (doesn't sound logical but still)
9 years ago
pictuga
82f01acfb1
Fix types (int mostly) in facebook
9 years ago
pictuga
fa14fa0ccb
Fix encoding in feedify
9 years ago
pictuga
d2345a94f8
Speed up a little html encoding detection
...
Not sure this is a smart move
9 years ago
pictuga
245ba99ae9
Always clean up downloads' enconding
...
TPB, fuck ye
9 years ago
pictuga
e3f525ff2a
Fix JSON related encoding issue
9 years ago
pictuga
c43bf9f35f
Feed: print roottree
...
Keeps doctype declaration, xml stylesheets and so on
9 years ago
pictuga
71e4d597f8
Atom: support for <link rel="relative">
...
Not very clean. Clean work would require full rewrite of xgetCreate & maybe the whole item "selection" system
9 years ago
pictuga
e2512dba0c
Cath all errors in threads
...
Makes in possible to empty the queue whatsoever happens
9 years ago
pictuga
1644a409ef
Add application/xml as html mimetype
9 years ago
pictuga
8b3062eaa2
Make Accept headers py 2.6 compatible
9 years ago
pictuga
66c6dcf203
Store "as it" in JSON (int or str)
...
The json lib does all the work
9 years ago
pictuga
52fe67beb3
Create cache file when missing
...
Damn it...
9 years ago
pictuga
020c095194
Pretty-print JSON to cache
...
Easier for debugging, not really needed though, takes more place for nothing
9 years ago
pictuga
c83f5bb23d
URI based Cache filename
...
Problem: lengthy filenames, have to cut them after 255 chars, sucks with FB, can cause problems
9 years ago
pictuga
1a50cb390a
User JSON to save cache
...
Faster, cleaner, safer, unobfuscated, not 1/3 bigger
9 years ago
pictuga
fde5d3e8c5
Argument "silent" prevents from outputting feed
...
Also fixes small indentation typo (non blocking though)
9 years ago
pictuga
1d10d30173
Reddit: check feed url instead of link url
9 years ago
pictuga
b927a4d089
Use "str.startswith" instead of ugly hack
9 years ago
pictuga
8916216403
Add "txt" argument, to change HTTP Content-Type
9 years ago
pictuga
336d98230f
Make sure final data is there in downloader
9 years ago
pictuga
2634dab40c
Use HTTP Accept headers
...
Tries to tell servers what we need, hopefull this will avoid the useless donwloading of pdf, images and so on (especially on Hacker News)
9 years ago
pictuga
0de5adc505
Remove return value in one __ini__
...
It was some kind of previous code left-over
9 years ago
pictuga
363014ee4c
User IOError to catch errors
...
Most http related exceptions inherit from that one
9 years ago
pictuga
7156dd5522
Support for non-textual gzipped http content
9 years ago
pictuga
ecc18dc4ab
Thread Gather()
...
Twice faster. Ctrl-C no longer works in terminal. Don't know how to fix it. Using threading.active_count() or sth didn't work as expected. New settings (THREADS) to change the number of threads to use.
9 years ago
pictuga
8832611a05
Fix MAX_ITEM condition
9 years ago
pictuga
1274df4e1d
Yet another feedburner workaround
...
This one really drove me mad. When the "Referer" is feedburner itself, it just will show html page (with Google Reader Add buttons), instead of the regular feed. Itself adding "?format=xml" at the end of the url also solves it, but users don't know about that...
9 years ago
pictuga
d3db04332e
Don't run Fix() on dropped elements
9 years ago
pictuga
35fba5ffad
Toggle isPermaLing when changing item url
9 years ago
pictuga
91251ec306
isPermaLink in feeds.py (RSS only)
9 years ago
