Yesterday

• 45 hours ago

  r1155 (missing comma) committed by hueniverse   -   missing comma

  missing comma
• 45 hours ago

  r1154 ([No log message]) committed by hueniverse   -   [No log message]

  [No log message]
• 47 hours ago

  r1153 ([No log message]) committed by hueniverse   -   [No log message]

  [No log message]
• 47 hours ago

  r1152 (Removed all HTML versions of the specs to prevent normative ...) committed by hueniverse   -   Removed all HTML versions of the specs to prevent normative references to working copies

  Removed all HTML versions of the specs to prevent normative references to working copies

Last 7 days

• Dec 17, 2009

  issue 138 (In C# OAuthBase.cs, UrlEncoded Paremeters (such as a search ...) reported by bryan.crow   -   What steps will reproduce the problem? In C# (haven't tested the other versions) Send an OAUTH request including a url encoded query parameter (such as a search string with it's spaces converted to "%20"s). The C# code currently double-url-encodes any already url-encoded query string parameters, resulting in an invalid signature base (ex: the %20 improperly becomes %2520). What is the expected output? What do you see instead? The signature is invalid because of the double-encoded reserved characters. Please use labels and text to provide additional information. I've uploaded a new version of OAuthBase.cs with a patch in place. Simply UrlDecode the normalizedRequestParameters before re-encoding them. This ensures there is no double-enoding. The change is on OAuthBase.cs line 274. - Bryan

  What steps will reproduce the problem? In C# (haven't tested the other versions) Send an OAUTH request including a url encoded query parameter (such as a search string with it's spaces converted to "%20"s). The C# code currently double-url-encodes any already url-encoded query string parameters, resulting in an invalid signature base (ex: the %20 improperly becomes %2520). What is the expected output? What do you see instead? The signature is invalid because of the double-encoded reserved characters. Please use labels and text to provide additional information. I've uploaded a new version of OAuthBase.cs with a patch in place. Simply UrlDecode the normalizedRequestParameters before re-encoding them. This ensures there is no double-enoding. The change is on OAuthBase.cs line 274. - Bryan

Last 30 days

• Dec 09, 2009

  issue 113 (Unit tests for Python lib) commented on by jroesslein   -   I have begun working on unit tests for the library. I'll post the patches here as they get made.

  I have begun working on unit tests for the library. I'll post the patches here as they get made.
• Dec 09, 2009

  issue 113 (Unit tests for Python lib) commented on by jroesslein   -   I have begun working on unit tests for the library. I will be posting the patch series here as them come.

  I have begun working on unit tests for the library. I will be posting the patch series here as them come.
• Dec 09, 2009

  issue 121 (get_callback_url() in OAuthToken doesn't append the token) commented on by jroesslein   -   Attached is a fix for this issue.

  Attached is a fix for this issue.
• Dec 09, 2009

  issue 121 (get_callback_url() in OAuthToken doesn't append the token) commented on by jroesslein   -   Attached is a patch for resolve this issue.

  Attached is a patch for resolve this issue.
• Dec 08, 2009

  issue 88 (Python API does not support multi-valued parameters in query...) commented on by jroesslein   -   Attached is a patch that resolves this issue. Github commit: http://github.com/joshthecoder/python-oauth/commit/18f8a6b985237fd14cb57528e6d2b58257f4ad2b

  Attached is a patch that resolves this issue. Github commit: http://github.com/joshthecoder/python-oauth/commit/18f8a6b985237fd14cb57528e6d2b58257f4ad2b
• Dec 07, 2009

  r1151 (draft-hammer-token-auth-00) committed by hueniverse   -   draft-hammer-token-auth-00

  draft-hammer-token-auth-00
• Dec 06, 2009

  r1150 (Interim save) committed by hueniverse   -   Interim save

  Interim save
• Dec 06, 2009

  r1149 (Initial and incomplete token scheme draft) committed by hueniverse   -   Initial and incomplete token scheme draft

  Initial and incomplete token scheme draft
• Dec 03, 2009

  r1148 (doc history) committed by hueniverse   -   doc history

  doc history
• Dec 03, 2009

  r1147 (Added 'with a 200 status code (OK)' in two places.) committed by hueniverse   -   Added 'with a 200 status code (OK)' in two places.

  Added 'with a 200 status code (OK)' in two places.
• Dec 03, 2009

  r1146 (draft-hammer-oauth-08) committed by hueniverse   -   draft-hammer-oauth-08

  draft-hammer-oauth-08
• Dec 02, 2009

  r1145 (Applied comments from IESG members) committed by hueniverse   -   Applied comments from IESG members

  Applied comments from IESG members
• Nov 25, 2009

  issue 137 (Incorrect TimeStamp on OAuthBase.cs) commented on by giancarlo.todone   -   To be correct, the fix should be applied both on OAuthBase and in OAuthUtils. The latter overrides GenerateTimeStamp() from OAuthBase trying to implement the same thing in a different way. return Convert.ToInt64(span.TotalSeconds).ToString(); is a correct implementation, while string str = span.TotalSeconds.ToString(); int index = str.IndexOf("."); ecc... is not. I suggest deleting OAuthUtil.GenerateTimeStamp() override.

  To be correct, the fix should be applied both on OAuthBase and in OAuthUtils. The latter overrides GenerateTimeStamp() from OAuthBase trying to implement the same thing in a different way. return Convert.ToInt64(span.TotalSeconds).ToString(); is a correct implementation, while string str = span.TotalSeconds.ToString(); int index = str.IndexOf("."); ecc... is not. I suggest deleting OAuthUtil.GenerateTimeStamp() override.
• Nov 25, 2009

  issue 137 (Incorrect TimeStamp on OAuthBase.cs) reported by giorgio.davanzo   -   As reported here: http://blog.stevienova.com/2008/04/19/oauth-getting-started-with-oauth-in-c- net/ GenerateTimeStamp() returns an incorrect timestamp, adding decimal points and hence generating an Invalid Signature response. The solution there provided, however, is not totally correct since it does not account for different locales---i.e., the decimal point separator is not always a ".", but could also be a ",".

  As reported here: http://blog.stevienova.com/2008/04/19/oauth-getting-started-with-oauth-in-c- net/ GenerateTimeStamp() returns an incorrect timestamp, adding decimal points and hence generating an Invalid Signature response. The solution there provided, however, is not totally correct since it does not account for different locales---i.e., the decimal point separator is not always a ".", but could also be a ",".
• Nov 23, 2009

  r1144 (draft-hammer-oauth-07) committed by hueniverse   -   draft-hammer-oauth-07

  draft-hammer-oauth-07

Earlier this year

• Nov 20, 2009

  r1143 (draft-hammer-oauth-06) committed by hueniverse   -   draft-hammer-oauth-06

  draft-hammer-oauth-06
• Nov 18, 2009

  issue 136 (lookup_nonce in abstract class OAuthDatastore of the Python ...) reported by sumeet.a   -   Either that or the example server's MockOAuthDataStore.lookup_nonce is wrong. The abstract class says to return OAuthToken while MockOAuthDataStore returns an str. It doesn't seem like the difference would really break anything, but I figured you'd want to know about the inconsistency.

  Either that or the example server's MockOAuthDataStore.lookup_nonce is wrong. The abstract class says to return OAuthToken while MockOAuthDataStore returns an str. It doesn't seem like the difference would really break anything, but I figured you'd want to know about the inconsistency.
• Nov 17, 2009

  issue 126 (Python library: OAuthRequest.to_postdata() and OAuthRequest....) commented on by scherpenisse   -   Would be nice if this gets fixed at some point... :) I attached the proposed patch.

  Would be nice if this gets fixed at some point... :) I attached the proposed patch.
• Nov 16, 2009

  r1142 (draft-hammer-oauth-05) committed by hueniverse   -   draft-hammer-oauth-05

  draft-hammer-oauth-05
• Nov 15, 2009

  r1141 (Removed EC draft) committed by hueniverse   -   Removed EC draft

  Removed EC draft
• Nov 15, 2009

  r1140 (-04 revision of 1.0) committed by hueniverse   -   -04 revision of 1.0

  -04 revision of 1.0
• Nov 13, 2009

  r1139 (Fixed a few things, see Changes ) committed by keith.grennan   -   Fixed a few things, see Changes

  Fixed a few things, see Changes
• Nov 12, 2009

  issue 58 (PHP: Suggestion: add sample code for accessing the API) commented on by robert.ganly   -   i agree this would be a great improvement, i hit a bit of a hurdle here as it isn't very intuitive, esp. if new to oauth.

  i agree this would be a great improvement, i hit a bit of a hurdle here as it isn't very intuitive, esp. if new to oauth.
• Nov 04, 2009

  issue 126 (Python library: OAuthRequest.to_postdata() and OAuthRequest....) commented on by ralphm.net   -   Confirming this issue and the fix proposed by violinwise. Interestingly enough, get_normalized_parameters does perform the required encoding to UTF-8.

  Confirming this issue and the fix proposed by violinwise. Interestingly enough, get_normalized_parameters does perform the required encoding to UTF-8.
• Nov 01, 2009

  r1138 (removing httpclient 4.0 beta 1 jar and httpcore 4.0 beta 2 j...) committed by s...@seansullivan.com   -   removing httpclient 4.0 beta 1 jar and httpcore 4.0 beta 2 jar.

  removing httpclient 4.0 beta 1 jar and httpcore 4.0 beta 2 jar.
• Oct 23, 2009

  r1137 (Changed svn:eol-style to native.) committed by jmkristian   -   Changed svn:eol-style to native.

  Changed svn:eol-style to native.
• Oct 23, 2009

  r1136 (Changed svn:eol-style to native.) committed by jmkristian   -   Changed svn:eol-style to native.

  Changed svn:eol-style to native.
• Oct 23, 2009

  r1135 (Handle URLs that contain '@' in the path part.) committed by jmkristian   -   Handle URLs that contain '@' in the path part.

  Handle URLs that contain '@' in the path part.
• Oct 22, 2009

  issue 135 (python client example code don't using httplib properly ) reported by Filia.Tao   -   line 62 : self.connection.request(oauth_request.http_method, self.request_token_url, headers=oauth_request.to_header()) second argument should be a path not the full url . GET http://xxx.xxxx.com/oauth/get_request_token HTTP/1.1 Host: xxx.xxxx.com Accept-Encoding: identity Authorization: OAuth realm="", oauth_nonce="50270268", oauth_timestamp="1256217898", oauth_consumer_key="some-key", oauth_signature_method="HMAC-SHA1", oauth_version="1.0", oauth_signature="Ttkf%2FRm%2BEq6HSb4C0qTKlBx5Ups%3D"' that's bad . should be GET /oauth/get_request_token HTTP/1.1

  line 62 : self.connection.request(oauth_request.http_method, self.request_token_url, headers=oauth_request.to_header()) second argument should be a path not the full url . GET http://xxx.xxxx.com/oauth/get_request_token HTTP/1.1 Host: xxx.xxxx.com Accept-Encoding: identity Authorization: OAuth realm="", oauth_nonce="50270268", oauth_timestamp="1256217898", oauth_consumer_key="some-key", oauth_signature_method="HMAC-SHA1", oauth_version="1.0", oauth_signature="Ttkf%2FRm%2BEq6HSb4C0qTKlBx5Ups%3D"' that's bad . should be GET /oauth/get_request_token HTTP/1.1
• Oct 20, 2009

  r1134 (Fix for big timestamps. Thanks Zed Shaw!) committed by leah.culver   -   Fix for big timestamps. Thanks Zed Shaw!

  Fix for big timestamps. Thanks Zed Shaw!
• Oct 20, 2009

  r1133 (Feedback from Lisa about removing 'server credentials' term) committed by hueniverse   -   Feedback from Lisa about removing 'server credentials' term

  Feedback from Lisa about removing 'server credentials' term
• Oct 19, 2009

  r1132 (Branch for JMeter plugin. ) committed by zhihong   -   Branch for JMeter plugin.

  Branch for JMeter plugin.
• Oct 14, 2009

  issue 134 (bug in OAToken.m , token encoded.) reported by mickael.laloum   -   There is a little bug in OAToken in the initWithHTTPResponseBody:body method, the token are encoded. As it say in http://oauth.net/core/1.0/#response_parameters SO de must decode it before affect to the attributs. it become : self.key = [[elements objectAtIndex:1] URLDecodedString] ; and self.secret = [[elements objectAtIndex:1] URLDecodedString]; don't forget the import NSString+URLEncoding.h.

  There is a little bug in OAToken in the initWithHTTPResponseBody:body method, the token are encoded. As it say in http://oauth.net/core/1.0/#response_parameters SO de must decode it before affect to the attributs. it become : self.key = [[elements objectAtIndex:1] URLDecodedString] ; and self.secret = [[elements objectAtIndex:1] URLDecodedString]; don't forget the import NSString+URLEncoding.h.
• Oct 12, 2009

  issue 133 (Python: Checking nonces needs access to the oauth_timestamp) reported by raumkraut   -   According to the OAuth spec; a nonce "is unique for all requests with that timestamp". However, OAuthDataStore.lookup_nonce is passed only the consumer, token, and nonce; none of which provide the relevant oauth_timestamp. Hence, using the library as it stands, the nonce cannot actually be validated as per the spec.

  According to the OAuth spec; a nonce "is unique for all requests with that timestamp". However, OAuthDataStore.lookup_nonce is passed only the consumer, token, and nonce; none of which provide the relevant oauth_timestamp. Hence, using the library as it stands, the nonce cannot actually be validated as per the spec.
• Oct 09, 2009

  issue 49 (Ampersands in signatures double encoded when making POST OAu...) commented on by morten.fangel   -   #4: There is a patch which solves it, hasn't made it into SVN.. When I get the time I will make sure it makes it's way to the official repo.. Sorry, I've been preoccupied with other duties..

  #4: There is a patch which solves it, hasn't made it into SVN.. When I get the time I will make sure it makes it's way to the official repo.. Sorry, I've been preoccupied with other duties..
• Oct 09, 2009

  issue 49 (Ampersands in signatures double encoded when making POST OAu...) commented on by adwo...@wagenheim.com   -   Did this ever get fixed? I'm having problems with POST's failing.

  Did this ever get fixed? I'm having problems with POST's failing.
• Oct 07, 2009

  issue 132 (Problem while running the js library outside of the context ...) reported by mepartoconmigo   -   What steps will reproduce the problem? 1. Run the library outside the context of a browser, like in Rhino js runtime What is the expected output? What do you see instead? Expected the sing of a request, but instead i got the error 'document object not found' What version of the product are you using? On what operating system? svn version of the library Please provide any additional information below. Seems the declaration of document.getElementById is failling from environments like Rhino. Thanks, Francisco

  What steps will reproduce the problem? 1. Run the library outside the context of a browser, like in Rhino js runtime What is the expected output? What do you see instead? Expected the sing of a request, but instead i got the error 'document object not found' What version of the product are you using? On what operating system? svn version of the library Please provide any additional information below. Seems the declaration of document.getElementById is failling from environments like Rhino. Thanks, Francisco
• Oct 05, 2009

  issue 131 ([PHP] Several tests broken) reported by ithinkihaveacat   -   On a fresh svn checkout (r1131) phpunit produces the following output: PHPUnit 3.4.0 by Sebastian Bergmann. .....F.............FF...F.F.....F.........FF........... Time: 0 seconds There were 8 failures: 1) OAuthRequestTest::testCreateRequestFromConsumerAndToken Failed asserting that two strings are equal. --- Expected +++ Actual @@ @@ -bar + /mnt/hgfs/workspace/php-oauth/tests/OAuthRequestTest.php:92 2) OAuthRequestTest::testBuildSignature Failed asserting that two strings are equal. --- Expected +++ Actual @@ @@ -kd94hf93k423kf44&pfkkdhi9sl3r4s00 +kd94hf93k423kf44%26pfkkdhi9sl3r4s00 /mnt/hgfs/workspace/php-oauth/tests/OAuthRequestTest.php:277 3) OAuthRequestTest::testSign Failed asserting that two strings are equal. --- Expected +++ Actual @@ @@ -kd94hf93k423kf44&pfkkdhi9sl3r4s00 +kd94hf93k423kf44%26pfkkdhi9sl3r4s00 /mnt/hgfs/workspace/php-oauth/tests/OAuthRequestTest.php:305 4) OAuthServerTest::testRejectRequestWithMissingParameters Allowed a request without `oauth_signature_method` /mnt/hgfs/workspace/php-oauth/tests/OAuthServerTest.php:78 5) OAuthServerTest::testRejectFutureTimestamp Expected exception OAuthException 6) OAuthServerTest::testRejectUnknownVersion Expected exception OAuthException 7) OAuthSignatureMethodPlaintextTest::testBuildSignature Failed asserting that two strings are equal. --- Expected +++ Actual @@ @@ -cs& +cs%26 /mnt/hgfs/workspace/php-oauth/tests/OAuthSignatureMethodPlaintextTest.php:22 8) OAuthSignatureMethodPlaintextTest::testVerifySignature Failed asserting that <boolean:false> is true. /mnt/hgfs/workspace/php-oauth/tests/OAuthSignatureMethodPlaintextTest.php:52 FAILURES! Tests: 55, Assertions: 139, Failures: 8.

  On a fresh svn checkout (r1131) phpunit produces the following output: PHPUnit 3.4.0 by Sebastian Bergmann. .....F.............FF...F.F.....F.........FF........... Time: 0 seconds There were 8 failures: 1) OAuthRequestTest::testCreateRequestFromConsumerAndToken Failed asserting that two strings are equal. --- Expected +++ Actual @@ @@ -bar + /mnt/hgfs/workspace/php-oauth/tests/OAuthRequestTest.php:92 2) OAuthRequestTest::testBuildSignature Failed asserting that two strings are equal. --- Expected +++ Actual @@ @@ -kd94hf93k423kf44&pfkkdhi9sl3r4s00 +kd94hf93k423kf44%26pfkkdhi9sl3r4s00 /mnt/hgfs/workspace/php-oauth/tests/OAuthRequestTest.php:277 3) OAuthRequestTest::testSign Failed asserting that two strings are equal. --- Expected +++ Actual @@ @@ -kd94hf93k423kf44&pfkkdhi9sl3r4s00 +kd94hf93k423kf44%26pfkkdhi9sl3r4s00 /mnt/hgfs/workspace/php-oauth/tests/OAuthRequestTest.php:305 4) OAuthServerTest::testRejectRequestWithMissingParameters Allowed a request without `oauth_signature_method` /mnt/hgfs/workspace/php-oauth/tests/OAuthServerTest.php:78 5) OAuthServerTest::testRejectFutureTimestamp Expected exception OAuthException 6) OAuthServerTest::testRejectUnknownVersion Expected exception OAuthException 7) OAuthSignatureMethodPlaintextTest::testBuildSignature Failed asserting that two strings are equal. --- Expected +++ Actual @@ @@ -cs& +cs%26 /mnt/hgfs/workspace/php-oauth/tests/OAuthSignatureMethodPlaintextTest.php:22 8) OAuthSignatureMethodPlaintextTest::testVerifySignature Failed asserting that <boolean:false> is true. /mnt/hgfs/workspace/php-oauth/tests/OAuthSignatureMethodPlaintextTest.php:52 FAILURES! Tests: 55, Assertions: 139, Failures: 8.
• Oct 05, 2009

  issue 130 (split() should be replaced with explode()) reported by ithinkihaveacat   -   The split() function http://php.net/split is deprecated as of PHP 5.3. It should be replaced with explode(). The patch is: Index: OAuth.php =========================================================== ======== --- OAuth.php (revision 1131) +++ OAuth.php (working copy) @@ -717,11 +717,11 @@ public static function parse_parameters( $input ) { if (!isset($input) || !$input) return array(); - $pairs = split('&', $input); + $pairs = explode('&', $input); $parsed_parameters = array(); foreach ($pairs as $pair) { - $split = split('=', $pair, 2); + $split = explode('=', $pair, 2); $parameter = OAuthUtil::urldecode_rfc3986($split[0]); $value = isset($split[1]) ? OAuthUtil::urldecode_rfc3986($split[1]) : '';

  The split() function http://php.net/split is deprecated as of PHP 5.3. It should be replaced with explode(). The patch is: Index: OAuth.php =========================================================== ======== --- OAuth.php (revision 1131) +++ OAuth.php (working copy) @@ -717,11 +717,11 @@ public static function parse_parameters( $input ) { if (!isset($input) || !$input) return array(); - $pairs = split('&', $input); + $pairs = explode('&', $input); $parsed_parameters = array(); foreach ($pairs as $pair) { - $split = split('=', $pair, 2); + $split = explode('=', $pair, 2); $parameter = OAuthUtil::urldecode_rfc3986($split[0]); $value = isset($split[1]) ? OAuthUtil::urldecode_rfc3986($split[1]) : '';
• Oct 05, 2009

  issue 129 (nonce timestamp returned when out of range returns in millis...) reported by jsummers2   -   What steps will reproduce the problem? 1. Send invalid timestamp in seconds 2. Server returns invalid timestamp but states valid range in milliseconds not seconds What is the expected output? What do you see instead? The spec states that the time should be submitted in seconds and not milliseconds but return error statement states milliseconds range as the valid input. What version of the product are you using? On what operating system? Java, 20090531, Debian Please provide any additional information below. Server returns : oauth_problem=timestamp_refused&oauth_acceptable_timestamps=1246561582853-1246562182853 But I think it should return a valid timestamp like this : oauth_problem=timestamp_refused&oauth_acceptable_timestamps=1246561582-1246562182

  What steps will reproduce the problem? 1. Send invalid timestamp in seconds 2. Server returns invalid timestamp but states valid range in milliseconds not seconds What is the expected output? What do you see instead? The spec states that the time should be submitted in seconds and not milliseconds but return error statement states milliseconds range as the valid input. What version of the product are you using? On what operating system? Java, 20090531, Debian Please provide any additional information below. Server returns : oauth_problem=timestamp_refused&oauth_acceptable_timestamps=1246561582853-1246562182853 But I think it should return a valid timestamp like this : oauth_problem=timestamp_refused&oauth_acceptable_timestamps=1246561582-1246562182
• Oct 04, 2009

  r1131 (upgraded oauth-consumer from 20090617 to 20090823) committed by s...@seansullivan.com   -   upgraded oauth-consumer from 20090617 to 20090823

  upgraded oauth-consumer from 20090617 to 20090823
• Oct 02, 2009

  issue 128 (Memory leak in [OAMutableURLRequest _signatureBaseString]) reported by jhender...@anu.ie   -   We have an app that incorporates the Obj-C 2.0 lib for Twitter OAuth. When we run the app through instruments we discover a memory leak tracing back to this method. We can't figure out why because everything seems to be autoreleased. The issue appears to be with the following: [self URLEncodedString: normalizedRequestParameters] which is part of: // OAuth Spec, Section 9.1.2 "Concatenate Request Elements" NSString *ret = [NSString stringWithFormat:@"%@&%@&%@", [self HTTPMethod], [self URLEncodedString: [self URLStringWithoutQueryFromURL: [self URL]]], [self URLEncodedString: normalizedRequestParameters]]; I can't work out what to do to fix this - we've already cleared one memory issue by releasing signature at the bottom of -(void)prepare in the same file, but this one has stumped us. When we get to the bottom of it, we'll post a patch, but any help is appreciated along the way. We're completely stuck! Thanks, John

  We have an app that incorporates the Obj-C 2.0 lib for Twitter OAuth. When we run the app through instruments we discover a memory leak tracing back to this method. We can't figure out why because everything seems to be autoreleased. The issue appears to be with the following: [self URLEncodedString: normalizedRequestParameters] which is part of: // OAuth Spec, Section 9.1.2 "Concatenate Request Elements" NSString *ret = [NSString stringWithFormat:@"%@&%@&%@", [self HTTPMethod], [self URLEncodedString: [self URLStringWithoutQueryFromURL: [self URL]]], [self URLEncodedString: normalizedRequestParameters]]; I can't work out what to do to fix this - we've already cleared one memory issue by releasing signature at the bottom of -(void)prepare in the same file, but this one has stumped us. When we get to the bottom of it, we'll post a patch, but any help is appreciated along the way. We're completely stuck! Thanks, John
• Oct 01, 2009

  issue 127 (Python: OAuthError definition mostly redundant) reported by raumkraut   -   Notwithstanding issues #110 and #123, the class definition of OAuthError seems to be pretty much entirely redundant. A simple `pass`ing class definition inheriting from Exception would serve exactly the same purpose, no? The only 'feature' I can see the current definition adding is a default message ("OAuth error occured"), which is already made entirely redundant by the name of the class itself - "OAuthError". It also has the spelling of "occurred" wrong. ;) I specifically suggest inheriting from Exception rather than the current RuntimeError, because RuntimeError "is mostly a relic from a previous version of the interpreter" (so sayeth the Python docs).

  Notwithstanding issues #110 and #123, the class definition of OAuthError seems to be pretty much entirely redundant. A simple `pass`ing class definition inheriting from Exception would serve exactly the same purpose, no? The only 'feature' I can see the current definition adding is a default message ("OAuth error occured"), which is already made entirely redundant by the name of the class itself - "OAuthError". It also has the spelling of "occurred" wrong. ;) I specifically suggest inheriting from Exception rather than the current RuntimeError, because RuntimeError "is mostly a relic from a previous version of the interpreter" (so sayeth the Python docs).
• Sep 30, 2009

  issue 126 (Python library: OAuthRequest.to_postdata() and OAuthRequest....) reported by violinwise   -   Simply run the attached script. I tried to fix it. In the definitions of the functions to_postdata() and to_header(), I replaced str() with _utf8_str(), and it worked.

  Simply run the attached script. I tried to fix it. In the definitions of the functions to_postdata() and to_header(), I replaced str() with _utf8_str(), and it worked.
• Sep 29, 2009

  r1130 (Python: changes to setup.py for version.) committed by leah.culver   -   Python: changes to setup.py for version.

  Python: changes to setup.py for version.