Re: Reviewboard 2.0.6 hungs on new request with non latin symbols
2.0.8 fixes my problem too ! Thanks. Le jeudi 28 août 2014 15:09:02 UTC+2, koval...@gmail.com a écrit : > > I am using Reviewboard 2.0.6 with Postgres database in utf-8. > When I making new pre-commit request on svn-based repository with > diff-file containing non-latin symbols, reviewboard hungs. > If diff file contains only latin symbols, it works ok. > > Is it possible to fix this bug? > -- Get the Review Board Power Pack at http://www.reviewboard.org/powerpack/ --- Sign up for Review Board hosting at RBCommons: https://rbcommons.com/ --- Happy user? Let us know at http://www.reviewboard.org/users/ --- You received this message because you are subscribed to the Google Groups "reviewboard" group. To unsubscribe from this group and stop receiving emails from it, send an email to reviewboard+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: Reviewboard 2.0.6 hungs on new request with non latin symbols
Glad it works, Sara! Thanks for your patience and feedback on this! Christian -- Christian Hammond - christ...@beanbaginc.com Review Board - https://www.reviewboard.org Beanbag, Inc. - https://www.beanbaginc.com On September 24, 2014 at 6:10:36 AM, Sara Rayburn (sararayb...@gmail.com) wrote: > 2.0.8 fixes this for us. Thanks so much! > > On Tuesday, September 23, 2014 6:32:55 PM UTC-5, Christian Hammond wrote: > > > > Hi Sara, > > > > David has a repro case for this now, so we're looking into a fix. > > > > There will be a 2.0.8 release tonight. > > > > Christian > > > > -- > > Christian Hammond - chri...@beanbaginc.com > > Review Board - https://www.reviewboard.org > > Beanbag, Inc. - https://www.beanbaginc.com > > > > > > On September 23, 2014 at 3:17:04 PM, Christian Hammond ( > > chri...@beanbaginc.com ) wrote: > > > Hi Sara, > > > > > > Something that would really help is a complete dump of the > > request/responses (headers > > > and payloads) for any requests to /api/* in the Network tab of Firefox > > when these issues > > > occur. > > > > > > I know some information was sent before, but those look to be the > > Console tab only. > > > > > > Christian > > > > > > -- > > > Christian Hammond - chri...@beanbaginc.com > > > Review Board - https://www.reviewboard.org > > > Beanbag, Inc. - https://www.beanbaginc.com > > > > > > > > > On September 23, 2014 at 2:23:34 PM, Christian Hammond ( > > chri...@beanbaginc.com ) > > > wrote: > > > > I wish I knew too... > > > > > > > > So, for each of these cases, is the only visible error the "rsp is > > null" error? > > > > > > > > This is tricky to diagnose without more direct access to the server. > > Hoping something > > > > will allow me to repro it here. > > > > > > > > Christian > > > > > > > > -- > > > > Christian Hammond - chri...@beanbaginc.com > > > > Review Board - https://www.reviewboard.org > > > > Beanbag, Inc. - https://www.beanbaginc.com > > > > > > > > > > > > On September 23, 2014 at 1:43:11 PM, Sara Rayburn (sarar...@gmail.com > > ) wrote: > > > > > I wish I knew what this was. We've had three developers run into > > this bug > > > > > today alone. It's becoming a real problem for us. More answers to > > your > > > > > questions inline below. > > > > > Sara > > > > > > > > > > On Tuesday, September 23, 2014 2:08:28 PM UTC-5, Christian Hammond > > wrote: > > > > > > > > > > > > Hi Sara, > > > > > > > > > > > > > > > > > > On September 23, 2014 at 11:39:53 AM, Sara Rayburn ( > > sarar...@gmail.com > > > > > > ) wrote: > > > > > > > On Tuesday, September 23, 2014 1:11:18 PM UTC-5, Christian > > Hammond > > > > > > wrote: > > > > > > > > > > > > > > > > Hi Sara, > > > > > > > > > > > > > > > > There are several things here that I am very confused about: > > > > > > > > > > > > > > > > 1) Can you clarify which part is new since 2.0.7? The subject > > says > > > > > > 2.0.6. > > > > > > > > > > > > > > > > > > > > > > That is weird. I wasn't the original poster. We have only seen > > the bug > > > > > > > since updating to 2.0.7. I know this because we have a request > > with > > > > > > > non-unicode contents that was uploaded before I updated, without > > > > > > problems. > > > > > > > Trying to reuse the same patch for an additional review request > > > > > > triggered > > > > > > > the same unicode error we saw previously. > > > > > > > > > > > > On September 9th, you posted saying "I am having a similar problem > > after > > > > > > upgrading to 2.0.6." We didn't have a 2.0.7 release until the > > 15th. > > > > > > > > > > > > > > > > > > Good point. I guess it's a problem since before that. > > > > > > > > > > > > > > > > > > > 1) Is this still only happening with Firefox? > > > > > > > > > > > > > > > Yes > > > > > > > > > > > > > > > > > > > > > > > 2) What version of Firefox? > > > > > > > > > > > > > > > > 30 > > > > > > > > > > > > Does 33 work? I haven't had any issues with any recent version... > > > > > > > > > > > > > > > > > > Tried in most recent Firefox (35), issue remains. > > > > > > > > > > > > > > > > > > 4) If MySQL, are the tables set for Unicode? > > > > > > > > > > > > > > > > The tables are a mix of latin1 and utf8. Perhaps this is the > > problem? > > > > > > > > > > > > This is likely to be the cause of at least some of your problems. > > You'll > > > > > > need to make sure they're all in utf8, otherwise there will be > > issues even > > > > > > populating the database with non-latin1 text. This can break us in > > numerous > > > > > > ways. > > > > > > > > > > > > I made sure the database encodings were utf8 and even set the > > encoding for > > > > > the repository in question to utf8 in the admin repo settings for > > > > > reviewboard. I added utf8 as the charset in settings_local.py. All > > of these > > > > > changes later, I still have the problem. > > > > > > > > > > > > > > > > > > > 5) Are you using any form of proxy server at your company? > > > > > > > > > > > > > > > > No. Very vanilla setup on a
Re: Reviewboard 2.0.6 hungs on new request with non latin symbols
2.0.8 fixes this for us. Thanks so much! On Tuesday, September 23, 2014 6:32:55 PM UTC-5, Christian Hammond wrote: > > Hi Sara, > > David has a repro case for this now, so we're looking into a fix. > > There will be a 2.0.8 release tonight. > > Christian > > -- > Christian Hammond - chri...@beanbaginc.com > Review Board - https://www.reviewboard.org > Beanbag, Inc. - https://www.beanbaginc.com > > > On September 23, 2014 at 3:17:04 PM, Christian Hammond ( > chri...@beanbaginc.com ) wrote: > > Hi Sara, > > > > Something that would really help is a complete dump of the > request/responses (headers > > and payloads) for any requests to /api/* in the Network tab of Firefox > when these issues > > occur. > > > > I know some information was sent before, but those look to be the > Console tab only. > > > > Christian > > > > -- > > Christian Hammond - chri...@beanbaginc.com > > Review Board - https://www.reviewboard.org > > Beanbag, Inc. - https://www.beanbaginc.com > > > > > > On September 23, 2014 at 2:23:34 PM, Christian Hammond ( > chri...@beanbaginc.com ) > > wrote: > > > I wish I knew too... > > > > > > So, for each of these cases, is the only visible error the "rsp is > null" error? > > > > > > This is tricky to diagnose without more direct access to the server. > Hoping something > > > will allow me to repro it here. > > > > > > Christian > > > > > > -- > > > Christian Hammond - chri...@beanbaginc.com > > > Review Board - https://www.reviewboard.org > > > Beanbag, Inc. - https://www.beanbaginc.com > > > > > > > > > On September 23, 2014 at 1:43:11 PM, Sara Rayburn (sarar...@gmail.com > ) wrote: > > > > I wish I knew what this was. We've had three developers run into > this bug > > > > today alone. It's becoming a real problem for us. More answers to > your > > > > questions inline below. > > > > Sara > > > > > > > > On Tuesday, September 23, 2014 2:08:28 PM UTC-5, Christian Hammond > wrote: > > > > > > > > > > Hi Sara, > > > > > > > > > > > > > > > On September 23, 2014 at 11:39:53 AM, Sara Rayburn ( > sarar...@gmail.com > > > > > ) wrote: > > > > > > On Tuesday, September 23, 2014 1:11:18 PM UTC-5, Christian > Hammond > > > > > wrote: > > > > > > > > > > > > > > Hi Sara, > > > > > > > > > > > > > > There are several things here that I am very confused about: > > > > > > > > > > > > > > 1) Can you clarify which part is new since 2.0.7? The subject > says > > > > > 2.0.6. > > > > > > > > > > > > > > > > > > > That is weird. I wasn't the original poster. We have only seen > the bug > > > > > > since updating to 2.0.7. I know this because we have a request > with > > > > > > non-unicode contents that was uploaded before I updated, without > > > > > problems. > > > > > > Trying to reuse the same patch for an additional review request > > > > > triggered > > > > > > the same unicode error we saw previously. > > > > > > > > > > On September 9th, you posted saying "I am having a similar problem > after > > > > > upgrading to 2.0.6." We didn't have a 2.0.7 release until the > 15th. > > > > > > > > > > > > > > > Good point. I guess it's a problem since before that. > > > > > > > > > > > > > > > > 1) Is this still only happening with Firefox? > > > > > > > > > > > > > Yes > > > > > > > > > > > > > > > > > > > > 2) What version of Firefox? > > > > > > > > > > > > > > 30 > > > > > > > > > > Does 33 work? I haven't had any issues with any recent version... > > > > > > > > > > > > > > > Tried in most recent Firefox (35), issue remains. > > > > > > > > > > > > > > > 4) If MySQL, are the tables set for Unicode? > > > > > > > > > > > > > > The tables are a mix of latin1 and utf8. Perhaps this is the > problem? > > > > > > > > > > This is likely to be the cause of at least some of your problems. > You'll > > > > > need to make sure they're all in utf8, otherwise there will be > issues even > > > > > populating the database with non-latin1 text. This can break us in > numerous > > > > > ways. > > > > > > > > > > I made sure the database encodings were utf8 and even set the > encoding for > > > > the repository in question to utf8 in the admin repo settings for > > > > reviewboard. I added utf8 as the charset in settings_local.py. All > of these > > > > changes later, I still have the problem. > > > > > > > > > > > > > > > > 5) Are you using any form of proxy server at your company? > > > > > > > > > > > > > > No. Very vanilla setup on a VM. > > > > > > > > > > How about HTTP proxies that the users configure in their browsers? > We've > > > > > seen cases in the past of some proxies messing with the responses > from our > > > > > API. > > > > > > > > > > Nope, no one. > > > > > > > > > Christian > > > > > > > > > > > > > > > -- > > > > > Christian Hammond - chri...@beanbaginc.com > > > > > Review Board - https://www.reviewboard.org > > > > > Beanbag, Inc. - https://www.beanb
Re: Reviewboard 2.0.6 hungs on new request with non latin symbols
Hi Sara, David has a repro case for this now, so we're looking into a fix. There will be a 2.0.8 release tonight. Christian -- Christian Hammond - christ...@beanbaginc.com Review Board - https://www.reviewboard.org Beanbag, Inc. - https://www.beanbaginc.com On September 23, 2014 at 3:17:04 PM, Christian Hammond (christ...@beanbaginc.com) wrote: > Hi Sara, > > Something that would really help is a complete dump of the request/responses > (headers > and payloads) for any requests to /api/* in the Network tab of Firefox when > these issues > occur. > > I know some information was sent before, but those look to be the Console tab > only. > > Christian > > -- > Christian Hammond - christ...@beanbaginc.com > Review Board - https://www.reviewboard.org > Beanbag, Inc. - https://www.beanbaginc.com > > > On September 23, 2014 at 2:23:34 PM, Christian Hammond > (christ...@beanbaginc.com) > wrote: > > I wish I knew too... > > > > So, for each of these cases, is the only visible error the "rsp is null" > > error? > > > > This is tricky to diagnose without more direct access to the server. Hoping > > something > > will allow me to repro it here. > > > > Christian > > > > -- > > Christian Hammond - christ...@beanbaginc.com > > Review Board - https://www.reviewboard.org > > Beanbag, Inc. - https://www.beanbaginc.com > > > > > > On September 23, 2014 at 1:43:11 PM, Sara Rayburn (sararayb...@gmail.com) > > wrote: > > > I wish I knew what this was. We've had three developers run into this bug > > > today alone. It's becoming a real problem for us. More answers to your > > > questions inline below. > > > Sara > > > > > > On Tuesday, September 23, 2014 2:08:28 PM UTC-5, Christian Hammond wrote: > > > > > > > > Hi Sara, > > > > > > > > > > > > On September 23, 2014 at 11:39:53 AM, Sara Rayburn (sarar...@gmail.com > > > > ) wrote: > > > > > On Tuesday, September 23, 2014 1:11:18 PM UTC-5, Christian Hammond > > > > wrote: > > > > > > > > > > > > Hi Sara, > > > > > > > > > > > > There are several things here that I am very confused about: > > > > > > > > > > > > 1) Can you clarify which part is new since 2.0.7? The subject says > > > > 2.0.6. > > > > > > > > > > > > > > > > That is weird. I wasn't the original poster. We have only seen the bug > > > > > since updating to 2.0.7. I know this because we have a request with > > > > > non-unicode contents that was uploaded before I updated, without > > > > problems. > > > > > Trying to reuse the same patch for an additional review request > > > > triggered > > > > > the same unicode error we saw previously. > > > > > > > > On September 9th, you posted saying "I am having a similar problem after > > > > upgrading to 2.0.6." We didn't have a 2.0.7 release until the 15th. > > > > > > > > > > > > Good point. I guess it's a problem since before that. > > > > > > > > > > > > > 1) Is this still only happening with Firefox? > > > > > > > > > > > Yes > > > > > > > > > > > > > > > > > 2) What version of Firefox? > > > > > > > > > > > > 30 > > > > > > > > Does 33 work? I haven't had any issues with any recent version... > > > > > > > > > > > > Tried in most recent Firefox (35), issue remains. > > > > > > > > > > > > 4) If MySQL, are the tables set for Unicode? > > > > > > > > > > > > The tables are a mix of latin1 and utf8. Perhaps this is the > > > > > > problem? > > > > > > > > This is likely to be the cause of at least some of your problems. You'll > > > > need to make sure they're all in utf8, otherwise there will be issues > > > > even > > > > populating the database with non-latin1 text. This can break us in > > > > numerous > > > > ways. > > > > > > > > I made sure the database encodings were utf8 and even set the encoding > > > > for > > > the repository in question to utf8 in the admin repo settings for > > > reviewboard. I added utf8 as the charset in settings_local.py. All of > > > these > > > changes later, I still have the problem. > > > > > > > > > > > > > 5) Are you using any form of proxy server at your company? > > > > > > > > > > > > No. Very vanilla setup on a VM. > > > > > > > > How about HTTP proxies that the users configure in their browsers? We've > > > > seen cases in the past of some proxies messing with the responses from > > > > our > > > > API. > > > > > > > > Nope, no one. > > > > > > > Christian > > > > > > > > > > > > -- > > > > Christian Hammond - chri...@beanbaginc.com > > > > Review Board - https://www.reviewboard.org > > > > Beanbag, Inc. - https://www.beanbaginc.com > > > > > > > > > > > > > > > > > > > > > > > > > > > > > Thanks, > > > > > > > > > > > > Christian > > > > > > > > > > > > -- > > > > > > Christian Hammond - chri...@beanbaginc.com > > > > > > Review Board - https://www.reviewboard.org > > > > > > Beanbag, Inc. - https://www.beanbaginc.com > > > > > > > > > > > > > > > > > > On September 23, 2014 at 10:57:14 AM, Sara Rayburn > > > > > > (sarar...@gmail.com > > > > > > ) wrote: > > > > > > > I can also confirm that
Re: Reviewboard 2.0.6 hungs on new request with non latin symbols
Hi Sara, Something that would really help is a complete dump of the request/responses (headers and payloads) for any requests to /api/* in the Network tab of Firefox when these issues occur. I know some information was sent before, but those look to be the Console tab only. Christian -- Christian Hammond - christ...@beanbaginc.com Review Board - https://www.reviewboard.org Beanbag, Inc. - https://www.beanbaginc.com On September 23, 2014 at 2:23:34 PM, Christian Hammond (christ...@beanbaginc.com) wrote: > I wish I knew too... > > So, for each of these cases, is the only visible error the "rsp is null" > error? > > This is tricky to diagnose without more direct access to the server. Hoping > something > will allow me to repro it here. > > Christian > > -- > Christian Hammond - christ...@beanbaginc.com > Review Board - https://www.reviewboard.org > Beanbag, Inc. - https://www.beanbaginc.com > > > On September 23, 2014 at 1:43:11 PM, Sara Rayburn (sararayb...@gmail.com) > wrote: > > I wish I knew what this was. We've had three developers run into this bug > > today alone. It's becoming a real problem for us. More answers to your > > questions inline below. > > Sara > > > > On Tuesday, September 23, 2014 2:08:28 PM UTC-5, Christian Hammond wrote: > > > > > > Hi Sara, > > > > > > > > > On September 23, 2014 at 11:39:53 AM, Sara Rayburn (sarar...@gmail.com > > > ) wrote: > > > > On Tuesday, September 23, 2014 1:11:18 PM UTC-5, Christian Hammond > > > wrote: > > > > > > > > > > Hi Sara, > > > > > > > > > > There are several things here that I am very confused about: > > > > > > > > > > 1) Can you clarify which part is new since 2.0.7? The subject says > > > 2.0.6. > > > > > > > > > > > > > That is weird. I wasn't the original poster. We have only seen the bug > > > > since updating to 2.0.7. I know this because we have a request with > > > > non-unicode contents that was uploaded before I updated, without > > > problems. > > > > Trying to reuse the same patch for an additional review request > > > triggered > > > > the same unicode error we saw previously. > > > > > > On September 9th, you posted saying "I am having a similar problem after > > > upgrading to 2.0.6." We didn't have a 2.0.7 release until the 15th. > > > > > > > > > Good point. I guess it's a problem since before that. > > > > > > > > > > 1) Is this still only happening with Firefox? > > > > > > > > > Yes > > > > > > > > > > > > > > 2) What version of Firefox? > > > > > > > > > > 30 > > > > > > Does 33 work? I haven't had any issues with any recent version... > > > > > > > > > Tried in most recent Firefox (35), issue remains. > > > > > > > > > 4) If MySQL, are the tables set for Unicode? > > > > > > > > > > The tables are a mix of latin1 and utf8. Perhaps this is the problem? > > > > > > This is likely to be the cause of at least some of your problems. You'll > > > need to make sure they're all in utf8, otherwise there will be issues even > > > populating the database with non-latin1 text. This can break us in > > > numerous > > > ways. > > > > > > I made sure the database encodings were utf8 and even set the encoding for > > the repository in question to utf8 in the admin repo settings for > > reviewboard. I added utf8 as the charset in settings_local.py. All of these > > changes later, I still have the problem. > > > > > > > > > > 5) Are you using any form of proxy server at your company? > > > > > > > > > > No. Very vanilla setup on a VM. > > > > > > How about HTTP proxies that the users configure in their browsers? We've > > > seen cases in the past of some proxies messing with the responses from our > > > API. > > > > > > Nope, no one. > > > > > Christian > > > > > > > > > -- > > > Christian Hammond - chri...@beanbaginc.com > > > Review Board - https://www.reviewboard.org > > > Beanbag, Inc. - https://www.beanbaginc.com > > > > > > > > > > > > > > > > > > > > > > > Thanks, > > > > > > > > > > Christian > > > > > > > > > > -- > > > > > Christian Hammond - chri...@beanbaginc.com > > > > > Review Board - https://www.reviewboard.org > > > > > Beanbag, Inc. - https://www.beanbaginc.com > > > > > > > > > > > > > > > On September 23, 2014 at 10:57:14 AM, Sara Rayburn (sarar...@gmail.com > > > > > ) wrote: > > > > > > I can also confirm that this is a new bug since 2.0.7. One of the > > > review > > > > > > requests giving us trouble was added before we updated, and when > > > someone > > > > > > tried to change the submitter on the request it seems to have run > > > into > > > > > the > > > > > > error again. When the same user copied the patch and created a new > > > > > review > > > > > > request (since they couldn't take ownership of the request) they ran > > > > > into > > > > > > the unicode bug again. > > > > > > > > > > > > On Tuesday, September 23, 2014 12:51:51 PM UTC-5, Sara Rayburn > > > wrote: > > > > > > > > > > > > > > Any new information on this bug? We've run into it several times. > > > > > > > > > > > > > > On T
Re: Reviewboard 2.0.6 hungs on new request with non latin symbols
I wish I knew too... So, for each of these cases, is the only visible error the "rsp is null" error? This is tricky to diagnose without more direct access to the server. Hoping something will allow me to repro it here. Christian -- Christian Hammond - christ...@beanbaginc.com Review Board - https://www.reviewboard.org Beanbag, Inc. - https://www.beanbaginc.com On September 23, 2014 at 1:43:11 PM, Sara Rayburn (sararayb...@gmail.com) wrote: > I wish I knew what this was. We've had three developers run into this bug > today alone. It's becoming a real problem for us. More answers to your > questions inline below. > Sara > > On Tuesday, September 23, 2014 2:08:28 PM UTC-5, Christian Hammond wrote: > > > > Hi Sara, > > > > > > On September 23, 2014 at 11:39:53 AM, Sara Rayburn (sarar...@gmail.com > > ) wrote: > > > On Tuesday, September 23, 2014 1:11:18 PM UTC-5, Christian Hammond > > wrote: > > > > > > > > Hi Sara, > > > > > > > > There are several things here that I am very confused about: > > > > > > > > 1) Can you clarify which part is new since 2.0.7? The subject says > > 2.0.6. > > > > > > > > > > That is weird. I wasn't the original poster. We have only seen the bug > > > since updating to 2.0.7. I know this because we have a request with > > > non-unicode contents that was uploaded before I updated, without > > problems. > > > Trying to reuse the same patch for an additional review request > > triggered > > > the same unicode error we saw previously. > > > > On September 9th, you posted saying "I am having a similar problem after > > upgrading to 2.0.6." We didn't have a 2.0.7 release until the 15th. > > > > > > Good point. I guess it's a problem since before that. > > > > > > > 1) Is this still only happening with Firefox? > > > > > > > Yes > > > > > > > > > > > 2) What version of Firefox? > > > > > > > > 30 > > > > Does 33 work? I haven't had any issues with any recent version... > > > > > > Tried in most recent Firefox (35), issue remains. > > > > > > 4) If MySQL, are the tables set for Unicode? > > > > > > > > The tables are a mix of latin1 and utf8. Perhaps this is the problem? > > > > This is likely to be the cause of at least some of your problems. You'll > > need to make sure they're all in utf8, otherwise there will be issues even > > populating the database with non-latin1 text. This can break us in numerous > > ways. > > > > I made sure the database encodings were utf8 and even set the encoding for > the repository in question to utf8 in the admin repo settings for > reviewboard. I added utf8 as the charset in settings_local.py. All of these > changes later, I still have the problem. > > > > > > > 5) Are you using any form of proxy server at your company? > > > > > > > > No. Very vanilla setup on a VM. > > > > How about HTTP proxies that the users configure in their browsers? We've > > seen cases in the past of some proxies messing with the responses from our > > API. > > > > Nope, no one. > > > Christian > > > > > > -- > > Christian Hammond - chri...@beanbaginc.com > > Review Board - https://www.reviewboard.org > > Beanbag, Inc. - https://www.beanbaginc.com > > > > > > > > > > > > > > > > > Thanks, > > > > > > > > Christian > > > > > > > > -- > > > > Christian Hammond - chri...@beanbaginc.com > > > > Review Board - https://www.reviewboard.org > > > > Beanbag, Inc. - https://www.beanbaginc.com > > > > > > > > > > > > On September 23, 2014 at 10:57:14 AM, Sara Rayburn (sarar...@gmail.com > > > > ) wrote: > > > > > I can also confirm that this is a new bug since 2.0.7. One of the > > review > > > > > requests giving us trouble was added before we updated, and when > > someone > > > > > tried to change the submitter on the request it seems to have run > > into > > > > the > > > > > error again. When the same user copied the patch and created a new > > > > review > > > > > request (since they couldn't take ownership of the request) they ran > > > > into > > > > > the unicode bug again. > > > > > > > > > > On Tuesday, September 23, 2014 12:51:51 PM UTC-5, Sara Rayburn > > wrote: > > > > > > > > > > > > Any new information on this bug? We've run into it several times. > > > > > > > > > > > > On Thursday, August 28, 2014 8:09:02 AM UTC-5, koval...@gmail.com > > > > wrote: > > > > > >> > > > > > >> I am using Reviewboard 2.0.6 with Postgres database in utf-8. > > > > > >> When I making new pre-commit request on svn-based repository with > > > > > >> diff-file containing non-latin symbols, reviewboard hungs. > > > > > >> If diff file contains only latin symbols, it works ok. > > > > > >> > > > > > >> Is it possible to fix this bug? > > > > > >> > > > > > > > > > > > > > > > > -- > > > > > Get the Review Board Power Pack at > > http://www.reviewboard.org/powerpack/ > > > > > --- > > > > > Sign up for Review Board hosting at RBCommons: > > https://rbcommons.com/ > > > > > --- > > > > > Happy user? Let us know at http://www.reviewboard.org/users/ > > > > > --- > > > > > You received this
Re: Reviewboard 2.0.6 hungs on new request with non latin symbols
I wish I knew what this was. We've had three developers run into this bug today alone. It's becoming a real problem for us. More answers to your questions inline below. Sara On Tuesday, September 23, 2014 2:08:28 PM UTC-5, Christian Hammond wrote: > > Hi Sara, > > > On September 23, 2014 at 11:39:53 AM, Sara Rayburn (sarar...@gmail.com > ) wrote: > > On Tuesday, September 23, 2014 1:11:18 PM UTC-5, Christian Hammond > wrote: > > > > > > Hi Sara, > > > > > > There are several things here that I am very confused about: > > > > > > 1) Can you clarify which part is new since 2.0.7? The subject says > 2.0.6. > > > > > > > That is weird. I wasn't the original poster. We have only seen the bug > > since updating to 2.0.7. I know this because we have a request with > > non-unicode contents that was uploaded before I updated, without > problems. > > Trying to reuse the same patch for an additional review request > triggered > > the same unicode error we saw previously. > > On September 9th, you posted saying "I am having a similar problem after > upgrading to 2.0.6." We didn't have a 2.0.7 release until the 15th. > > > Good point. I guess it's a problem since before that. > > > > 1) Is this still only happening with Firefox? > > > > > Yes > > > > > > > > 2) What version of Firefox? > > > > > > 30 > > Does 33 work? I haven't had any issues with any recent version... > > > Tried in most recent Firefox (35), issue remains. > > > 4) If MySQL, are the tables set for Unicode? > > > > > > The tables are a mix of latin1 and utf8. Perhaps this is the problem? > > This is likely to be the cause of at least some of your problems. You'll > need to make sure they're all in utf8, otherwise there will be issues even > populating the database with non-latin1 text. This can break us in numerous > ways. > > I made sure the database encodings were utf8 and even set the encoding for the repository in question to utf8 in the admin repo settings for reviewboard. I added utf8 as the charset in settings_local.py. All of these changes later, I still have the problem. > > > > 5) Are you using any form of proxy server at your company? > > > > > > No. Very vanilla setup on a VM. > > How about HTTP proxies that the users configure in their browsers? We've > seen cases in the past of some proxies messing with the responses from our > API. > > Nope, no one. > Christian > > > -- > Christian Hammond - chri...@beanbaginc.com > Review Board - https://www.reviewboard.org > Beanbag, Inc. - https://www.beanbaginc.com > > > > > > > > > > > Thanks, > > > > > > Christian > > > > > > -- > > > Christian Hammond - chri...@beanbaginc.com > > > Review Board - https://www.reviewboard.org > > > Beanbag, Inc. - https://www.beanbaginc.com > > > > > > > > > On September 23, 2014 at 10:57:14 AM, Sara Rayburn (sarar...@gmail.com > > > ) wrote: > > > > I can also confirm that this is a new bug since 2.0.7. One of the > review > > > > requests giving us trouble was added before we updated, and when > someone > > > > tried to change the submitter on the request it seems to have run > into > > > the > > > > error again. When the same user copied the patch and created a new > > > review > > > > request (since they couldn't take ownership of the request) they ran > > > into > > > > the unicode bug again. > > > > > > > > On Tuesday, September 23, 2014 12:51:51 PM UTC-5, Sara Rayburn > wrote: > > > > > > > > > > Any new information on this bug? We've run into it several times. > > > > > > > > > > On Thursday, August 28, 2014 8:09:02 AM UTC-5, koval...@gmail.com > > > wrote: > > > > >> > > > > >> I am using Reviewboard 2.0.6 with Postgres database in utf-8. > > > > >> When I making new pre-commit request on svn-based repository with > > > > >> diff-file containing non-latin symbols, reviewboard hungs. > > > > >> If diff file contains only latin symbols, it works ok. > > > > >> > > > > >> Is it possible to fix this bug? > > > > >> > > > > > > > > > > > > > -- > > > > Get the Review Board Power Pack at > http://www.reviewboard.org/powerpack/ > > > > --- > > > > Sign up for Review Board hosting at RBCommons: > https://rbcommons.com/ > > > > --- > > > > Happy user? Let us know at http://www.reviewboard.org/users/ > > > > --- > > > > You received this message because you are subscribed to the Google > > > Groups "reviewboard" > > > > group. > > > > To unsubscribe from this group and stop receiving emails from it, > send > > > an email to reviewboard...@googlegroups.com . > > > > For more options, visit https://groups.google.com/d/optout. > > > > > > > > > > > > > > -- > > Get the Review Board Power Pack at http://www.reviewboard.org/powerpack/ > > --- > > Sign up for Review Board hosting at RBCommons: https://rbcommons.com/ > > --- > > Happy user? Let us know at http://www.reviewboard.org/users/ > > --- > > You received this mess
Re: Reviewboard 2.0.6 hungs on new request with non latin symbols
Hi Sara, On September 23, 2014 at 11:39:53 AM, Sara Rayburn (sararayb...@gmail.com) wrote: > On Tuesday, September 23, 2014 1:11:18 PM UTC-5, Christian Hammond wrote: > > > > Hi Sara, > > > > There are several things here that I am very confused about: > > > > 1) Can you clarify which part is new since 2.0.7? The subject says 2.0.6. > > > > That is weird. I wasn't the original poster. We have only seen the bug > since updating to 2.0.7. I know this because we have a request with > non-unicode contents that was uploaded before I updated, without problems. > Trying to reuse the same patch for an additional review request triggered > the same unicode error we saw previously. On September 9th, you posted saying "I am having a similar problem after upgrading to 2.0.6." We didn't have a 2.0.7 release until the 15th. > > 1) Is this still only happening with Firefox? > > > Yes > > > > > 2) What version of Firefox? > > > > 30 Does 33 work? I haven't had any issues with any recent version... > > 4) If MySQL, are the tables set for Unicode? > > > > The tables are a mix of latin1 and utf8. Perhaps this is the problem? This is likely to be the cause of at least some of your problems. You'll need to make sure they're all in utf8, otherwise there will be issues even populating the database with non-latin1 text. This can break us in numerous ways. > > 5) Are you using any form of proxy server at your company? > > > > No. Very vanilla setup on a VM. How about HTTP proxies that the users configure in their browsers? We've seen cases in the past of some proxies messing with the responses from our API. Christian -- Christian Hammond - christ...@beanbaginc.com Review Board - https://www.reviewboard.org Beanbag, Inc. - https://www.beanbaginc.com > > > > > Thanks, > > > > Christian > > > > -- > > Christian Hammond - chri...@beanbaginc.com > > Review Board - https://www.reviewboard.org > > Beanbag, Inc. - https://www.beanbaginc.com > > > > > > On September 23, 2014 at 10:57:14 AM, Sara Rayburn (sarar...@gmail.com > > ) wrote: > > > I can also confirm that this is a new bug since 2.0.7. One of the review > > > requests giving us trouble was added before we updated, and when someone > > > tried to change the submitter on the request it seems to have run into > > the > > > error again. When the same user copied the patch and created a new > > review > > > request (since they couldn't take ownership of the request) they ran > > into > > > the unicode bug again. > > > > > > On Tuesday, September 23, 2014 12:51:51 PM UTC-5, Sara Rayburn wrote: > > > > > > > > Any new information on this bug? We've run into it several times. > > > > > > > > On Thursday, August 28, 2014 8:09:02 AM UTC-5, koval...@gmail.com > > wrote: > > > >> > > > >> I am using Reviewboard 2.0.6 with Postgres database in utf-8. > > > >> When I making new pre-commit request on svn-based repository with > > > >> diff-file containing non-latin symbols, reviewboard hungs. > > > >> If diff file contains only latin symbols, it works ok. > > > >> > > > >> Is it possible to fix this bug? > > > >> > > > > > > > > > > -- > > > Get the Review Board Power Pack at http://www.reviewboard.org/powerpack/ > > > --- > > > Sign up for Review Board hosting at RBCommons: https://rbcommons.com/ > > > --- > > > Happy user? Let us know at http://www.reviewboard.org/users/ > > > --- > > > You received this message because you are subscribed to the Google > > Groups "reviewboard" > > > group. > > > To unsubscribe from this group and stop receiving emails from it, send > > an email to reviewboard...@googlegroups.com . > > > For more options, visit https://groups.google.com/d/optout. > > > > > > > > > -- > Get the Review Board Power Pack at http://www.reviewboard.org/powerpack/ > --- > Sign up for Review Board hosting at RBCommons: https://rbcommons.com/ > --- > Happy user? Let us know at http://www.reviewboard.org/users/ > --- > You received this message because you are subscribed to the Google Groups > "reviewboard" > group. > To unsubscribe from this group and stop receiving emails from it, send an > email to reviewboard+unsubscr...@googlegroups.com. > For more options, visit https://groups.google.com/d/optout. > -- Get the Review Board Power Pack at http://www.reviewboard.org/powerpack/ --- Sign up for Review Board hosting at RBCommons: https://rbcommons.com/ --- Happy user? Let us know at http://www.reviewboard.org/users/ --- You received this message because you are subscribed to the Google Groups "reviewboard" group. To unsubscribe from this group and stop receiving emails from it, send an email to reviewboard+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: Reviewboard 2.0.6 hungs on new request with non latin symbols
On Tuesday, September 23, 2014 1:11:18 PM UTC-5, Christian Hammond wrote: > > Hi Sara, > > There are several things here that I am very confused about: > > 1) Can you clarify which part is new since 2.0.7? The subject says 2.0.6. > That is weird. I wasn't the original poster. We have only seen the bug since updating to 2.0.7. I know this because we have a request with non-unicode contents that was uploaded before I updated, without problems. Trying to reuse the same patch for an additional review request triggered the same unicode error we saw previously. > > 2) How are you changing the submitter on a review request? We don't have > any support for this, without modifying the database. > > We are in fact modifying the database via the admin tab. > 3) The original bug report seemed to have to do with JavaScript errors on > uploading diffs, but now it sounds like it's other fields as well? > > We haven't been able to come up with any working repro case for a diff > causing a hang. If that's the true cause of your hangs, we're definitely > going to need source files and patch files that consistently reproduce this > for our own testing. > I will try to provide a broken diff. > > Can you also answer a few more questions: > > 1) Is this still only happening with Firefox? > Yes > > 2) What version of Firefox? > > 30 > 3) What database server are you using, and which version? > > mysql 14.14 > 4) If MySQL, are the tables set for Unicode? > > The tables are a mix of latin1 and utf8. Perhaps this is the problem? > 5) Are you using any form of proxy server at your company? > > No. Very vanilla setup on a VM. > Thanks, > > Christian > > -- > Christian Hammond - chri...@beanbaginc.com > Review Board - https://www.reviewboard.org > Beanbag, Inc. - https://www.beanbaginc.com > > > On September 23, 2014 at 10:57:14 AM, Sara Rayburn (sarar...@gmail.com > ) wrote: > > I can also confirm that this is a new bug since 2.0.7. One of the review > > requests giving us trouble was added before we updated, and when someone > > tried to change the submitter on the request it seems to have run into > the > > error again. When the same user copied the patch and created a new > review > > request (since they couldn't take ownership of the request) they ran > into > > the unicode bug again. > > > > On Tuesday, September 23, 2014 12:51:51 PM UTC-5, Sara Rayburn wrote: > > > > > > Any new information on this bug? We've run into it several times. > > > > > > On Thursday, August 28, 2014 8:09:02 AM UTC-5, koval...@gmail.com > wrote: > > >> > > >> I am using Reviewboard 2.0.6 with Postgres database in utf-8. > > >> When I making new pre-commit request on svn-based repository with > > >> diff-file containing non-latin symbols, reviewboard hungs. > > >> If diff file contains only latin symbols, it works ok. > > >> > > >> Is it possible to fix this bug? > > >> > > > > > > > -- > > Get the Review Board Power Pack at http://www.reviewboard.org/powerpack/ > > --- > > Sign up for Review Board hosting at RBCommons: https://rbcommons.com/ > > --- > > Happy user? Let us know at http://www.reviewboard.org/users/ > > --- > > You received this message because you are subscribed to the Google > Groups "reviewboard" > > group. > > To unsubscribe from this group and stop receiving emails from it, send > an email to reviewboard...@googlegroups.com . > > For more options, visit https://groups.google.com/d/optout. > > > > -- Get the Review Board Power Pack at http://www.reviewboard.org/powerpack/ --- Sign up for Review Board hosting at RBCommons: https://rbcommons.com/ --- Happy user? Let us know at http://www.reviewboard.org/users/ --- You received this message because you are subscribed to the Google Groups "reviewboard" group. To unsubscribe from this group and stop receiving emails from it, send an email to reviewboard+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: Reviewboard 2.0.6 hungs on new request with non latin symbols
Hi Sara, There are several things here that I am very confused about: 1) Can you clarify which part is new since 2.0.7? The subject says 2.0.6. 2) How are you changing the submitter on a review request? We don't have any support for this, without modifying the database. 3) The original bug report seemed to have to do with JavaScript errors on uploading diffs, but now it sounds like it's other fields as well? We haven't been able to come up with any working repro case for a diff causing a hang. If that's the true cause of your hangs, we're definitely going to need source files and patch files that consistently reproduce this for our own testing. Can you also answer a few more questions: 1) Is this still only happening with Firefox? 2) What version of Firefox? 3) What database server are you using, and which version? 4) If MySQL, are the tables set for Unicode? 5) Are you using any form of proxy server at your company? Thanks, Christian -- Christian Hammond - christ...@beanbaginc.com Review Board - https://www.reviewboard.org Beanbag, Inc. - https://www.beanbaginc.com On September 23, 2014 at 10:57:14 AM, Sara Rayburn (sararayb...@gmail.com) wrote: > I can also confirm that this is a new bug since 2.0.7. One of the review > requests giving us trouble was added before we updated, and when someone > tried to change the submitter on the request it seems to have run into the > error again. When the same user copied the patch and created a new review > request (since they couldn't take ownership of the request) they ran into > the unicode bug again. > > On Tuesday, September 23, 2014 12:51:51 PM UTC-5, Sara Rayburn wrote: > > > > Any new information on this bug? We've run into it several times. > > > > On Thursday, August 28, 2014 8:09:02 AM UTC-5, koval...@gmail.com wrote: > >> > >> I am using Reviewboard 2.0.6 with Postgres database in utf-8. > >> When I making new pre-commit request on svn-based repository with > >> diff-file containing non-latin symbols, reviewboard hungs. > >> If diff file contains only latin symbols, it works ok. > >> > >> Is it possible to fix this bug? > >> > > > > -- > Get the Review Board Power Pack at http://www.reviewboard.org/powerpack/ > --- > Sign up for Review Board hosting at RBCommons: https://rbcommons.com/ > --- > Happy user? Let us know at http://www.reviewboard.org/users/ > --- > You received this message because you are subscribed to the Google Groups > "reviewboard" > group. > To unsubscribe from this group and stop receiving emails from it, send an > email to reviewboard+unsubscr...@googlegroups.com. > For more options, visit https://groups.google.com/d/optout. > -- Get the Review Board Power Pack at http://www.reviewboard.org/powerpack/ --- Sign up for Review Board hosting at RBCommons: https://rbcommons.com/ --- Happy user? Let us know at http://www.reviewboard.org/users/ --- You received this message because you are subscribed to the Google Groups "reviewboard" group. To unsubscribe from this group and stop receiving emails from it, send an email to reviewboard+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: Reviewboard 2.0.6 hungs on new request with non latin symbols
I can also confirm that this is a new bug since 2.0.7. One of the review requests giving us trouble was added before we updated, and when someone tried to change the submitter on the request it seems to have run into the error again. When the same user copied the patch and created a new review request (since they couldn't take ownership of the request) they ran into the unicode bug again. On Tuesday, September 23, 2014 12:51:51 PM UTC-5, Sara Rayburn wrote: > > Any new information on this bug? We've run into it several times. > > On Thursday, August 28, 2014 8:09:02 AM UTC-5, koval...@gmail.com wrote: >> >> I am using Reviewboard 2.0.6 with Postgres database in utf-8. >> When I making new pre-commit request on svn-based repository with >> diff-file containing non-latin symbols, reviewboard hungs. >> If diff file contains only latin symbols, it works ok. >> >> Is it possible to fix this bug? >> > -- Get the Review Board Power Pack at http://www.reviewboard.org/powerpack/ --- Sign up for Review Board hosting at RBCommons: https://rbcommons.com/ --- Happy user? Let us know at http://www.reviewboard.org/users/ --- You received this message because you are subscribed to the Google Groups "reviewboard" group. To unsubscribe from this group and stop receiving emails from it, send an email to reviewboard+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: Reviewboard 2.0.6 hungs on new request with non latin symbols
Any new information on this bug? We've run into it several times. On Thursday, August 28, 2014 8:09:02 AM UTC-5, koval...@gmail.com wrote: > > I am using Reviewboard 2.0.6 with Postgres database in utf-8. > When I making new pre-commit request on svn-based repository with > diff-file containing non-latin symbols, reviewboard hungs. > If diff file contains only latin symbols, it works ok. > > Is it possible to fix this bug? > -- Get the Review Board Power Pack at http://www.reviewboard.org/powerpack/ --- Sign up for Review Board hosting at RBCommons: https://rbcommons.com/ --- Happy user? Let us know at http://www.reviewboard.org/users/ --- You received this message because you are subscribed to the Google Groups "reviewboard" group. To unsubscribe from this group and stop receiving emails from it, send an email to reviewboard+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: Reviewboard 2.0.6 hungs on new request with non latin symbols
Hello, I have a similar issue, I found this in apache error.log : [Fri Sep 12 13:19:54.182670 2014] [:error] [pid 1339] WARNING:py.warnings:/usr/local/lib/python2.7/dist-packages/ReviewBoard-2.0.6-py2.7.egg/reviewboard/diffviewer/diffutils.py:81: UnicodeWarning: Unicode equal comparison failed to convert both arguments to Unicode - interpreting them as being unequal [Fri Sep 12 13:19:54.182690 2014] [:error] [pid 1339] if data == "": [Fri Sep 12 13:19:54.182697 2014] [:error] [pid 1339] Fatal Python error: PyEval_AcquireThread: NULL new thread state Le jeudi 28 août 2014 15:09:02 UTC+2, koval...@gmail.com a écrit : > > I am using Reviewboard 2.0.6 with Postgres database in utf-8. > When I making new pre-commit request on svn-based repository with > diff-file containing non-latin symbols, reviewboard hungs. > If diff file contains only latin symbols, it works ok. > > Is it possible to fix this bug? > -- Get the Review Board Power Pack at http://www.reviewboard.org/powerpack/ --- Sign up for Review Board hosting at RBCommons: https://rbcommons.com/ --- Happy user? Let us know at http://www.reviewboard.org/users/ --- You received this message because you are subscribed to the Google Groups "reviewboard" group. To unsubscribe from this group and stop receiving emails from it, send an email to reviewboard+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: Reviewboard 2.0.6 hungs on new request with non latin symbols
Hi David,
Oddly enough, there is nothing at all in the logs when I try to create the
failed new review request. When I do the same operations in Chrome (where
the diff works just fine), I see
2014-09-11 12:57:32,499 - DEBUG - - DiffParser.parse: Beginning parse of
diff, size = 5408
2014-09-11 12:57:32,503 - DEBUG - - DiffParser.parse: Finished parsing
diff.
2014-09-11 12:57:32,653 - DEBUG - - DiffParser.parse: Beginning parse of
diff, size = 5408
2014-09-11 12:57:32,656 - DEBUG - - DiffParser.parse: Finished parsing
diff.
in the logs. Nothing at all hits the logs using firefox. There are no
errors in the apache logs either.
Sara
On Wednesday, September 10, 2014 4:03:39 PM UTC-5, David Trowbridge wrote:
>
> Sara,
>
> Can you look in the server log and see if there are any tracebacks in
> there?
>
> -David
>
> On Wed, Sep 10, 2014 at 5:50 AM, Sara Rayburn > wrote:
>
>>
>> Christian,
>>
>> Yes, that was the output from the network tab for the failing new review
>> request, and the tab was open the whole time. When I create new review
>> request for that repository with one of the diffs that works, I get these
>> requests (I may not have caught them all... not familiar with web
>> debugging):
>> GET http://reviewboard.cctechnol.com/r/4078/ [HTTP/1.1 200 OK 197ms]
>> GET http://reviewboard.cctechnol.com/jsi18n/ [HTTP/1.1 200 OK 68ms]
>> GET http://reviewboard.cctechnol.com/api/review-requests/4078/ [HTTP/1.1
>> 200 OK 109ms]
>> GET
>> http://reviewboard.cctechnol.com/static/rb/css/common.min.e6e71be831db.css
>> [HTTP/1.1 200 OK 12ms]
>> GET
>> http://reviewboard.cctechnol.com/static/rb/css/reviews.min.f35ccb97b383.css
>>
>> We were finally able to narrow down which were the offending characters.
>> The developer had copied some documentation from a pdf document, and it had
>> some non-ascii quote marks in it. He corrected the files with the following
>> perl:
>> perl -pi -e 's/[[:^ascii:]]//g' filename
>>
>> Hope that helps.
>>
>> Sara
>>
>> On Tuesday, September 9, 2014 4:33:06 PM UTC-5, Christian Hammond wrote:
>>>
>>> Hi Sara,
>>>
>>> Strange.. Is that output from the Network tab? Was that tab open when
>>> initially making the request?
>>>
>>> Christian
>>>
>>> --
>>> Christian Hammond - chri...@beanbaginc.com
>>> Review Board - https://www.reviewboard.org
>>> Beanbag, Inc. - https://www.beanbaginc.com
>>>
>>>
>>> On September 9, 2014 at 2:30:26 PM, Sara Rayburn (sarar...@gmail.com)
>>> wrote:
>>> > Hi Christian,
>>> >
>>> > Well, as it happens, the error only occurs while using Firefox (v.
>>> 30). The
>>> > user having troubles was relieved to find Chrome as a temporary
>>> workaround.
>>> > I was able to unminify the code, but it didn't point me to a specific
>>> line.
>>> > Maybe a firefox thing? I looked for a 500 error, but I don't really
>>> even
>>> > see a relevant looking request. Here is the log in question, but
>>> there's
>>> > not much to it. Clicking on the type error gets the big nasty block
>>> from
>>> > before.
>>> >
>>> > GET http://www.gravatar.com/avatar/e3a6e9470514467a04cbc94da028458e
>>> > [HTTP/1.1 304 Not Modified 796ms]
>>> > GET
>>> > http://reviewboard.cctechnol.com/static/rb/images/icons.
>>> 2eed2f830119.png
>>> > [HTTP/1.1 304 Not Modified 6ms]
>>> > GET
>>> > http://reviewboard.cctechnol.com/static/rb/images/logo@2x.
>>> 4e25cc3cacef.png
>>> > [HTTP/1.1 304 Not Modified 6ms]
>>> > Use of getPreventDefault() is deprecated. Use defaultPrevented
>>> instead.
>>> > jquery-1.8.2.min.cfa9051cc0b0.js:2
>>> > GET
>>> > http://reviewboard.cctechnol.com/static/rb/images/spinner.
>>> 73e57937304d.gif
>>> > [HTTP/1.1 304 Not Modified 5ms]
>>> > The non-standard sendAsBinary method is deprecated and will soon be
>>> > removed. Use the standard send(Blob data) method instead.
>>> > jquery-1.8.2.min.cfa9051cc0b0.js:2
>>> > TypeError: rsp is null
>>> >
>>> > On Tuesday, September 9, 2014 4:09:29 PM UTC-5, Christian Hammond
>>> wrote:
>>> > >
>>> > > Hi Sara,
>>> > >
>>> > > We minify our files by default, making the context less than useful.
>>> If
>>> > > you’re using Chrome, you should be able to click the error, see that
>>> big
>>> > > minified file, and then click the little “{}” at the bottom of the
>>> window.
>>> > > That will un-minify it, updating the error in the console to point
>>> to the
>>> > > actual line where things failed. If you could then paste that line
>>> (and the
>>> > > lines surrounding it), that will help narrow things down.
>>> > >
>>> > > We are getting ready for Review Board 2.0.7, which will display more
>>> > > useful errors in the New Review Request page when things go wrong.
>>> > > Hopefully it will help with this case as well.
>>> > >
>>> > > It would also be helpful to see the request/response payloads
>>> invoked by
>>> > > that page, in order to see where things went wrong. I’d expect one
>>> of the
>>> > > responses will be a HTTP 500
Re: Reviewboard 2.0.6 hungs on new request with non latin symbols
Sara,
Can you look in the server log and see if there are any tracebacks in there?
-David
On Wed, Sep 10, 2014 at 5:50 AM, Sara Rayburn wrote:
>
> Christian,
>
> Yes, that was the output from the network tab for the failing new review
> request, and the tab was open the whole time. When I create new review
> request for that repository with one of the diffs that works, I get these
> requests (I may not have caught them all... not familiar with web
> debugging):
> GET http://reviewboard.cctechnol.com/r/4078/ [HTTP/1.1 200 OK 197ms]
> GET http://reviewboard.cctechnol.com/jsi18n/ [HTTP/1.1 200 OK 68ms]
> GET http://reviewboard.cctechnol.com/api/review-requests/4078/ [HTTP/1.1
> 200 OK 109ms]
> GET
> http://reviewboard.cctechnol.com/static/rb/css/common.min.e6e71be831db.css
> [HTTP/1.1 200 OK 12ms]
> GET
> http://reviewboard.cctechnol.com/static/rb/css/reviews.min.f35ccb97b383.css
>
> We were finally able to narrow down which were the offending characters.
> The developer had copied some documentation from a pdf document, and it had
> some non-ascii quote marks in it. He corrected the files with the following
> perl:
> perl -pi -e 's/[[:^ascii:]]//g' filename
>
> Hope that helps.
>
> Sara
>
> On Tuesday, September 9, 2014 4:33:06 PM UTC-5, Christian Hammond wrote:
>>
>> Hi Sara,
>>
>> Strange.. Is that output from the Network tab? Was that tab open when
>> initially making the request?
>>
>> Christian
>>
>> --
>> Christian Hammond - chri...@beanbaginc.com
>> Review Board - https://www.reviewboard.org
>> Beanbag, Inc. - https://www.beanbaginc.com
>>
>>
>> On September 9, 2014 at 2:30:26 PM, Sara Rayburn (sarar...@gmail.com)
>> wrote:
>> > Hi Christian,
>> >
>> > Well, as it happens, the error only occurs while using Firefox (v. 30).
>> The
>> > user having troubles was relieved to find Chrome as a temporary
>> workaround.
>> > I was able to unminify the code, but it didn't point me to a specific
>> line.
>> > Maybe a firefox thing? I looked for a 500 error, but I don't really
>> even
>> > see a relevant looking request. Here is the log in question, but
>> there's
>> > not much to it. Clicking on the type error gets the big nasty block
>> from
>> > before.
>> >
>> > GET http://www.gravatar.com/avatar/e3a6e9470514467a04cbc94da028458e
>> > [HTTP/1.1 304 Not Modified 796ms]
>> > GET
>> > http://reviewboard.cctechnol.com/static/rb/images/icons.
>> 2eed2f830119.png
>> > [HTTP/1.1 304 Not Modified 6ms]
>> > GET
>> > http://reviewboard.cctechnol.com/static/rb/images/logo@2x.
>> 4e25cc3cacef.png
>> > [HTTP/1.1 304 Not Modified 6ms]
>> > Use of getPreventDefault() is deprecated. Use defaultPrevented instead.
>> > jquery-1.8.2.min.cfa9051cc0b0.js:2
>> > GET
>> > http://reviewboard.cctechnol.com/static/rb/images/spinner.
>> 73e57937304d.gif
>> > [HTTP/1.1 304 Not Modified 5ms]
>> > The non-standard sendAsBinary method is deprecated and will soon be
>> > removed. Use the standard send(Blob data) method instead.
>> > jquery-1.8.2.min.cfa9051cc0b0.js:2
>> > TypeError: rsp is null
>> >
>> > On Tuesday, September 9, 2014 4:09:29 PM UTC-5, Christian Hammond
>> wrote:
>> > >
>> > > Hi Sara,
>> > >
>> > > We minify our files by default, making the context less than useful.
>> If
>> > > you’re using Chrome, you should be able to click the error, see that
>> big
>> > > minified file, and then click the little “{}” at the bottom of the
>> window.
>> > > That will un-minify it, updating the error in the console to point to
>> the
>> > > actual line where things failed. If you could then paste that line
>> (and the
>> > > lines surrounding it), that will help narrow things down.
>> > >
>> > > We are getting ready for Review Board 2.0.7, which will display more
>> > > useful errors in the New Review Request page when things go wrong.
>> > > Hopefully it will help with this case as well.
>> > >
>> > > It would also be helpful to see the request/response payloads invoked
>> by
>> > > that page, in order to see where things went wrong. I’d expect one of
>> the
>> > > responses will be a HTTP 500.
>> > >
>> > > Christian
>> > >
>> > > --
>> > > Christian Hammond - chri...@beanbaginc.com
>> > > Review Board - https://www.reviewboard.org
>> > > Beanbag, Inc. - https://www.beanbaginc.com
>> > >
>> > >
>> >
>> > --
>> > Get the Review Board Power Pack at http://www.reviewboard.org/
>> powerpack/
>> > ---
>> > Sign up for Review Board hosting at RBCommons: https://rbcommons.com/
>> > ---
>> > Happy user? Let us know at http://www.reviewboard.org/users/
>> > ---
>> > You received this message because you are subscribed to the Google
>> Groups "reviewboard"
>> > group.
>> > To unsubscribe from this group and stop receiving emails from it, send
>> an email to reviewboard...@googlegroups.com.
>> > For more options, visit https://groups.google.com/d/optout.
>> >
>>
>> --
> Get the Review Board Power Pack at http://www.reviewboard.org/powerpack/
> ---
> Sign up for Review Board hosting at RBCommons: https://rbcommons.com/
> ---
> Happy user? Le
Re: Reviewboard 2.0.6 hungs on new request with non latin symbols
Christian,
Yes, that was the output from the network tab for the failing new review
request, and the tab was open the whole time. When I create new review
request for that repository with one of the diffs that works, I get these
requests (I may not have caught them all... not familiar with web
debugging):
GET http://reviewboard.cctechnol.com/r/4078/ [HTTP/1.1 200 OK 197ms]
GET http://reviewboard.cctechnol.com/jsi18n/ [HTTP/1.1 200 OK 68ms]
GET http://reviewboard.cctechnol.com/api/review-requests/4078/ [HTTP/1.1
200 OK 109ms]
GET
http://reviewboard.cctechnol.com/static/rb/css/common.min.e6e71be831db.css
[HTTP/1.1 200 OK 12ms]
GET
http://reviewboard.cctechnol.com/static/rb/css/reviews.min.f35ccb97b383.css
We were finally able to narrow down which were the offending characters.
The developer had copied some documentation from a pdf document, and it had
some non-ascii quote marks in it. He corrected the files with the following
perl:
perl -pi -e 's/[[:^ascii:]]//g' filename
Hope that helps.
Sara
On Tuesday, September 9, 2014 4:33:06 PM UTC-5, Christian Hammond wrote:
>
> Hi Sara,
>
> Strange.. Is that output from the Network tab? Was that tab open when
> initially making the request?
>
> Christian
>
> --
> Christian Hammond - chri...@beanbaginc.com
> Review Board - https://www.reviewboard.org
> Beanbag, Inc. - https://www.beanbaginc.com
>
>
> On September 9, 2014 at 2:30:26 PM, Sara Rayburn (sarar...@gmail.com
> ) wrote:
> > Hi Christian,
> >
> > Well, as it happens, the error only occurs while using Firefox (v. 30).
> The
> > user having troubles was relieved to find Chrome as a temporary
> workaround.
> > I was able to unminify the code, but it didn't point me to a specific
> line.
> > Maybe a firefox thing? I looked for a 500 error, but I don't really even
> > see a relevant looking request. Here is the log in question, but there's
> > not much to it. Clicking on the type error gets the big nasty block from
> > before.
> >
> > GET http://www.gravatar.com/avatar/e3a6e9470514467a04cbc94da028458e
> > [HTTP/1.1 304 Not Modified 796ms]
> > GET
> > http://reviewboard.cctechnol.com/static/rb/images/icons.2eed2f830119.png
>
> > [HTTP/1.1 304 Not Modified 6ms]
> > GET
> >
> http://reviewboard.cctechnol.com/static/rb/images/l...@2x.4e25cc3cacef.png
>
> > [HTTP/1.1 304 Not Modified 6ms]
> > Use of getPreventDefault() is deprecated. Use defaultPrevented instead.
> > jquery-1.8.2.min.cfa9051cc0b0.js:2
> > GET
> >
> http://reviewboard.cctechnol.com/static/rb/images/spinner.73e57937304d.gif
>
> > [HTTP/1.1 304 Not Modified 5ms]
> > The non-standard sendAsBinary method is deprecated and will soon be
> > removed. Use the standard send(Blob data) method instead.
> > jquery-1.8.2.min.cfa9051cc0b0.js:2
> > TypeError: rsp is null
> >
> > On Tuesday, September 9, 2014 4:09:29 PM UTC-5, Christian Hammond wrote:
> > >
> > > Hi Sara,
> > >
> > > We minify our files by default, making the context less than useful.
> If
> > > you’re using Chrome, you should be able to click the error, see that
> big
> > > minified file, and then click the little “{}” at the bottom of the
> window.
> > > That will un-minify it, updating the error in the console to point to
> the
> > > actual line where things failed. If you could then paste that line
> (and the
> > > lines surrounding it), that will help narrow things down.
> > >
> > > We are getting ready for Review Board 2.0.7, which will display more
> > > useful errors in the New Review Request page when things go wrong.
> > > Hopefully it will help with this case as well.
> > >
> > > It would also be helpful to see the request/response payloads invoked
> by
> > > that page, in order to see where things went wrong. I’d expect one of
> the
> > > responses will be a HTTP 500.
> > >
> > > Christian
> > >
> > > --
> > > Christian Hammond - chri...@beanbaginc.com
> > > Review Board - https://www.reviewboard.org
> > > Beanbag, Inc. - https://www.beanbaginc.com
> > >
> > >
> >
> > --
> > Get the Review Board Power Pack at http://www.reviewboard.org/powerpack/
> > ---
> > Sign up for Review Board hosting at RBCommons: https://rbcommons.com/
> > ---
> > Happy user? Let us know at http://www.reviewboard.org/users/
> > ---
> > You received this message because you are subscribed to the Google
> Groups "reviewboard"
> > group.
> > To unsubscribe from this group and stop receiving emails from it, send
> an email to reviewboard...@googlegroups.com .
> > For more options, visit https://groups.google.com/d/optout.
> >
>
>
--
Get the Review Board Power Pack at http://www.reviewboard.org/powerpack/
---
Sign up for Review Board hosting at RBCommons: https://rbcommons.com/
---
Happy user? Let us know at http://www.reviewboard.org/users/
---
You received this message because you are subscribed to the Google Groups
"reviewboard" group.
To unsubscribe from this group and stop re
Re: Reviewboard 2.0.6 hungs on new request with non latin symbols
Hi Sara,
Strange.. Is that output from the Network tab? Was that tab open when initially
making the request?
Christian
--
Christian Hammond - christ...@beanbaginc.com
Review Board - https://www.reviewboard.org
Beanbag, Inc. - https://www.beanbaginc.com
On September 9, 2014 at 2:30:26 PM, Sara Rayburn (sararayb...@gmail.com) wrote:
> Hi Christian,
>
> Well, as it happens, the error only occurs while using Firefox (v. 30). The
> user having troubles was relieved to find Chrome as a temporary workaround.
> I was able to unminify the code, but it didn't point me to a specific line.
> Maybe a firefox thing? I looked for a 500 error, but I don't really even
> see a relevant looking request. Here is the log in question, but there's
> not much to it. Clicking on the type error gets the big nasty block from
> before.
>
> GET http://www.gravatar.com/avatar/e3a6e9470514467a04cbc94da028458e
> [HTTP/1.1 304 Not Modified 796ms]
> GET
> http://reviewboard.cctechnol.com/static/rb/images/icons.2eed2f830119.png
> [HTTP/1.1 304 Not Modified 6ms]
> GET
> http://reviewboard.cctechnol.com/static/rb/images/l...@2x.4e25cc3cacef.png
> [HTTP/1.1 304 Not Modified 6ms]
> Use of getPreventDefault() is deprecated. Use defaultPrevented instead.
> jquery-1.8.2.min.cfa9051cc0b0.js:2
> GET
> http://reviewboard.cctechnol.com/static/rb/images/spinner.73e57937304d.gif
> [HTTP/1.1 304 Not Modified 5ms]
> The non-standard sendAsBinary method is deprecated and will soon be
> removed. Use the standard send(Blob data) method instead.
> jquery-1.8.2.min.cfa9051cc0b0.js:2
> TypeError: rsp is null
>
> On Tuesday, September 9, 2014 4:09:29 PM UTC-5, Christian Hammond wrote:
> >
> > Hi Sara,
> >
> > We minify our files by default, making the context less than useful. If
> > you’re using Chrome, you should be able to click the error, see that big
> > minified file, and then click the little “{}” at the bottom of the window.
> > That will un-minify it, updating the error in the console to point to the
> > actual line where things failed. If you could then paste that line (and the
> > lines surrounding it), that will help narrow things down.
> >
> > We are getting ready for Review Board 2.0.7, which will display more
> > useful errors in the New Review Request page when things go wrong.
> > Hopefully it will help with this case as well.
> >
> > It would also be helpful to see the request/response payloads invoked by
> > that page, in order to see where things went wrong. I’d expect one of the
> > responses will be a HTTP 500.
> >
> > Christian
> >
> > --
> > Christian Hammond - chri...@beanbaginc.com
> > Review Board - https://www.reviewboard.org
> > Beanbag, Inc. - https://www.beanbaginc.com
> >
> >
>
> --
> Get the Review Board Power Pack at http://www.reviewboard.org/powerpack/
> ---
> Sign up for Review Board hosting at RBCommons: https://rbcommons.com/
> ---
> Happy user? Let us know at http://www.reviewboard.org/users/
> ---
> You received this message because you are subscribed to the Google Groups
> "reviewboard"
> group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to reviewboard+unsubscr...@googlegroups.com.
> For more options, visit https://groups.google.com/d/optout.
>
--
Get the Review Board Power Pack at http://www.reviewboard.org/powerpack/
---
Sign up for Review Board hosting at RBCommons: https://rbcommons.com/
---
Happy user? Let us know at http://www.reviewboard.org/users/
---
You received this message because you are subscribed to the Google Groups
"reviewboard" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to reviewboard+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
Re: Reviewboard 2.0.6 hungs on new request with non latin symbols
Hi Christian,
Well, as it happens, the error only occurs while using Firefox (v. 30). The
user having troubles was relieved to find Chrome as a temporary workaround.
I was able to unminify the code, but it didn't point me to a specific line.
Maybe a firefox thing? I looked for a 500 error, but I don't really even
see a relevant looking request. Here is the log in question, but there's
not much to it. Clicking on the type error gets the big nasty block from
before.
GET http://www.gravatar.com/avatar/e3a6e9470514467a04cbc94da028458e
[HTTP/1.1 304 Not Modified 796ms]
GET
http://reviewboard.cctechnol.com/static/rb/images/icons.2eed2f830119.png
[HTTP/1.1 304 Not Modified 6ms]
GET
http://reviewboard.cctechnol.com/static/rb/images/l...@2x.4e25cc3cacef.png
[HTTP/1.1 304 Not Modified 6ms]
Use of getPreventDefault() is deprecated. Use defaultPrevented instead.
jquery-1.8.2.min.cfa9051cc0b0.js:2
GET
http://reviewboard.cctechnol.com/static/rb/images/spinner.73e57937304d.gif
[HTTP/1.1 304 Not Modified 5ms]
The non-standard sendAsBinary method is deprecated and will soon be
removed. Use the standard send(Blob data) method instead.
jquery-1.8.2.min.cfa9051cc0b0.js:2
TypeError: rsp is null
On Tuesday, September 9, 2014 4:09:29 PM UTC-5, Christian Hammond wrote:
>
> Hi Sara,
>
> We minify our files by default, making the context less than useful. If
> you’re using Chrome, you should be able to click the error, see that big
> minified file, and then click the little “{}” at the bottom of the window.
> That will un-minify it, updating the error in the console to point to the
> actual line where things failed. If you could then paste that line (and the
> lines surrounding it), that will help narrow things down.
>
> We are getting ready for Review Board 2.0.7, which will display more
> useful errors in the New Review Request page when things go wrong.
> Hopefully it will help with this case as well.
>
> It would also be helpful to see the request/response payloads invoked by
> that page, in order to see where things went wrong. I’d expect one of the
> responses will be a HTTP 500.
>
> Christian
>
> --
> Christian Hammond - chri...@beanbaginc.com
> Review Board - https://www.reviewboard.org
> Beanbag, Inc. - https://www.beanbaginc.com
>
>
--
Get the Review Board Power Pack at http://www.reviewboard.org/powerpack/
---
Sign up for Review Board hosting at RBCommons: https://rbcommons.com/
---
Happy user? Let us know at http://www.reviewboard.org/users/
---
You received this message because you are subscribed to the Google Groups
"reviewboard" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to reviewboard+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
Re: Reviewboard 2.0.6 hungs on new request with non latin symbols
Hi Sara,
We minify our files by default, making the context less than useful. If you’re
using Chrome, you should be able to click the error, see that big minified
file, and then click the little “{}” at the bottom of the window. That will
un-minify it, updating the error in the console to point to the actual line
where things failed. If you could then paste that line (and the lines
surrounding it), that will help narrow things down.
We are getting ready for Review Board 2.0.7, which will display more useful
errors in the New Review Request page when things go wrong. Hopefully it will
help with this case as well.
It would also be helpful to see the request/response payloads invoked by that
page, in order to see where things went wrong. I’d expect one of the responses
will be a HTTP 500.
Christian
--
Christian Hammond - christ...@beanbaginc.com
Review Board - https://www.reviewboard.org
Beanbag, Inc. - https://www.beanbaginc.com
On September 9, 2014 at 2:04:22 PM, Sara Rayburn (sararayb...@gmail.com) wrote:
> I am having a similar problem after upgrading to 2.0.6. I do see an error
> in my console for url http://reviewboard.cctechnol.com/r/new/. The error
> happens after I select a file from browse and try to upload.
>
> TypeError: rsp is null
>
> with the following details:
>
> (function(){RB.PostCommitModel=Backbone.Model.extend({defaults:{repository:null,branch:null}});RB.PreCommitModel=Backbone.Model.extend({defaults:{basedir:null,changeNumber:null,diffFile:null,diffValid:false,error:null,parentDiffFile:null,repository:null,state:0},State:{PROMPT_FOR_DIFF:0,PROMPT_FOR_BASEDIR:1,PROPMT_FOR_CHANGE_NUMBER:2,PROCESSING_DIFF:3,UPLOADING:4,PROMPT_FOR_PARENT_DIFF:5,ERROR:6},initialize:function(){_super(this).initialize.apply(this,arguments);this.on("change:diffFile
>
> change:parentDiffFile change:basedir "+"change:changeNumber
> change:diffValid",this._updateState,this)},startOver:function(){this.set({basedir:null,changeNumber:null,diffFile:null,diffValid:false,error:null,parentDiffFile:null,state:this.State.PROMPT_FOR_DIFF})},_updateState:function(){var
>
> basedir=this.get("basedir"),changeNumber=this.get("changeNumber"),diff=this.get("diffFile"),diffValid=this.get("diffValid"),parentDiff=this.get("parentDiffFile"),repository=this.get("repository"),requiresBasedir=repository.get("requiresBasedir"),requiresChangeNumber=repository.get("requiresChangeNumber"),state=this.get("state");switch(state){case
>
> this.State.PROMPT_FOR_DIFF:if(diff){if(requiresBasedir&&!basedir){this.set("state",this.State.PROMPT_FOR_BASEDIR)}else
>
> if(requiresChangeNumber&&!changeNumber){this.set("state",this.State.PROMPT_FOR_CHANGE_NUMBER)}else{this.set("state",this.State.PROCESSING_DIFF);this._tryValidate()}}break;case
>
> this.State.PROMPT_FOR_PARENT_DIFF:if(diff&&parentDiff){this.set("state",this.State.PROCESSING_DIFF);this._tryValidate()}break;case
>
> this.State.PROMPT_FOR_BASEDIR:console.assert(diff,"cannot be in basedir
> prompt
> state without a
> diff");if(basedir){if(requiresChangeNumber&&!changeNumber){this.set("state",this.State.PROMPT_FOR_CHANGE_NUMBER)}else{this.set("state",this.State.PROCESSING_DIFF);this._tryValidate()}}break;case
>
> this.State.PROMPT_FOR_CHANGE_NUMBER:console.assert(diff,"cannot be in
> changenum
> prompt state without a
> diff");if(changeNumber){this.set("state",this.State.PROCESSING_DIFF);this._tryValidate()}break;case
>
> this.State.PROCESSING_DIFF:if(diffValid){this.set("state",this.State.UPLOADING);this._createReviewRequest()}break;case
>
> this.State.UPLOADING:break;case
> this.State.ERROR:break}},_tryValidate:function(){var
> diff=this.get("diffFile"),parentDiff=this.get("parentDiffFile"),repository=this.get("repository"),uploader=new
>
> RB.ValidateDiffModel;this.set("diffValid",false);console.assert(diff);uploader.set({repository:repository.get("id"),localSitePrefix:repository.get("localSitePrefix"),basedir:this.get("basedir"),diff:diff,parentDiff:parentDiff});uploader.save({success:_.bind(this._onValidateSuccess,this),error:_.bind(this._onValidateError,this)})},_onValidateSuccess:function(){this.set("diffValid",true)},_onValidateError:function(model,xhr){var
>
> rsp=$.parseJSON(xhr.responseText),newState=this.State.ERROR,error;switch(rsp.err.code){case
>
> RB.APIErrors.REPO_FILE_NOT_FOUND:if(this.get("repository").get("scmtoolName")==="Git"&&rsp.revision.length!==40){error=gettext("The
>
> uploaded diff uses short revisions, but Review Board requires full revisions.
Please
> generate a new diff using the --full-index
> parameter.")}else{error=interpolate(gettext('The
> file "%s" (revision %s) was not found in the
> repository.'),[rsp.file,rsp.revision]);if(this.get("parentDiffFile")===null){newState=this.State.PROMPT_FOR_PARENT_DIFF}}break;case
>
> RB.APIErrors.DIFF_PARSE_ERROR:error=rsp.err.msg+"
"+"Line "+rsp.linenum+":
> "+rsp.reason;break;default:error=rsp.err.msg;break}if(error){this.set({state:ne
Re: Reviewboard 2.0.6 hungs on new request with non latin symbols
I am having a similar problem after upgrading to 2.0.6. I do see an error
in my console for url http://reviewboard.cctechnol.com/r/new/. The error
happens after I select a file from browse and try to upload.
TypeError: rsp is null
with the following details:
(function(){RB.PostCommitModel=Backbone.Model.extend({defaults:{repository:null,branch:null}});RB.PreCommitModel=Backbone.Model.extend({defaults:{basedir:null,changeNumber:null,diffFile:null,diffValid:false,error:null,parentDiffFile:null,repository:null,state:0},State:{PROMPT_FOR_DIFF:0,PROMPT_FOR_BASEDIR:1,PROPMT_FOR_CHANGE_NUMBER:2,PROCESSING_DIFF:3,UPLOADING:4,PROMPT_FOR_PARENT_DIFF:5,ERROR:6},initialize:function(){_super(this).initialize.apply(this,arguments);this.on("change:diffFile
change:parentDiffFile change:basedir "+"change:changeNumber
change:diffValid",this._updateState,this)},startOver:function(){this.set({basedir:null,changeNumber:null,diffFile:null,diffValid:false,error:null,parentDiffFile:null,state:this.State.PROMPT_FOR_DIFF})},_updateState:function(){var
basedir=this.get("basedir"),changeNumber=this.get("changeNumber"),diff=this.get("diffFile"),diffValid=this.get("diffValid"),parentDiff=this.get("parentDiffFile"),repository=this.get("repository"),requiresBasedir=repository.get("requiresBasedir"),requiresChangeNumber=repository.get("requiresChangeNumber"),state=this.get("state");switch(state){case
this.State.PROMPT_FOR_DIFF:if(diff){if(requiresBasedir&&!basedir){this.set("state",this.State.PROMPT_FOR_BASEDIR)}else
if(requiresChangeNumber&&!changeNumber){this.set("state",this.State.PROMPT_FOR_CHANGE_NUMBER)}else{this.set("state",this.State.PROCESSING_DIFF);this._tryValidate()}}break;case
this.State.PROMPT_FOR_PARENT_DIFF:if(diff&&parentDiff){this.set("state",this.State.PROCESSING_DIFF);this._tryValidate()}break;case
this.State.PROMPT_FOR_BASEDIR:console.assert(diff,"cannot be in basedir prompt
state without a
diff");if(basedir){if(requiresChangeNumber&&!changeNumber){this.set("state",this.State.PROMPT_FOR_CHANGE_NUMBER)}else{this.set("state",this.State.PROCESSING_DIFF);this._tryValidate()}}break;case
this.State.PROMPT_FOR_CHANGE_NUMBER:console.assert(diff,"cannot be in
changenum prompt state without a
diff");if(changeNumber){this.set("state",this.State.PROCESSING_DIFF);this._tryValidate()}break;case
this.State.PROCESSING_DIFF:if(diffValid){this.set("state",this.State.UPLOADING);this._createReviewRequest()}break;case
this.State.UPLOADING:break;case
this.State.ERROR:break}},_tryValidate:function(){var
diff=this.get("diffFile"),parentDiff=this.get("parentDiffFile"),repository=this.get("repository"),uploader=new
RB.ValidateDiffModel;this.set("diffValid",false);console.assert(diff);uploader.set({repository:repository.get("id"),localSitePrefix:repository.get("localSitePrefix"),basedir:this.get("basedir"),diff:diff,parentDiff:parentDiff});uploader.save({success:_.bind(this._onValidateSuccess,this),error:_.bind(this._onValidateError,this)})},_onValidateSuccess:function(){this.set("diffValid",true)},_onValidateError:function(model,xhr){var
rsp=$.parseJSON(xhr.responseText),newState=this.State.ERROR,error;switch(rsp.err.code){case
RB.APIErrors.REPO_FILE_NOT_FOUND:if(this.get("repository").get("scmtoolName")==="Git"&&rsp.revision.length!==40){error=gettext("The
uploaded diff uses short revisions, but Review Board requires full
revisions.Please generate a new diff using the --full-index
parameter.")}else{error=interpolate(gettext('The file "%s" (revision %s) was
not found in the
repository.'),[rsp.file,rsp.revision]);if(this.get("parentDiffFile")===null){newState=this.State.PROMPT_FOR_PARENT_DIFF}}break;case
RB.APIErrors.DIFF_PARSE_ERROR:error=rsp.err.msg+""+"Line
"+rsp.linenum+":
"+rsp.reason;break;default:error=rsp.err.msg;break}if(error){this.set({state:newState,error:error})}},_createReviewRequest:function(){var
repository=this.get("repository"),reviewRequest=new
RB.ReviewRequest({commitID:this.get("changeNumber"),localSitePrefix:repository.get("localSitePrefix"),repository:repository.get("id")});reviewRequest.save({success:function(){var
diff=reviewRequest.createDiff();diff.set({basedir:this.get("basedir"),diff:this.get("diffFile"),parentDiff:this.get("parentDiffFile")});diff.url=reviewRequest.get("links").diffs.href;diff.save({success:function(){window.location=reviewRequest.get("reviewURL")},error:this._onValidateError},this)},error:this._onValidateError},this)}});RB.NewReviewRequest=Backbone.Model.extend({defaults:{repositories:[]}});RB.CollectionView=Backbone.View.extend({itemViewType:null,initialize:function(options){var
collection=options.collection;this.collection=collection;this.views=[];collection.each(this._add,this);collection.on({add:this._add,remove:this._remove},this)},render:function(){this._rendered=true;this.$el.empty();_.each(this.views,function(view){this.$el.append(view.render().el)},this);return
this},_add:function(item){var
view;console.assert(this.itemViewType,"itemViewTyp
Re: Reviewboard 2.0.6 hungs on new request with non latin symbols
Yes, I use some extentions in Firifox, but with turned off extentions situation is the same. I don't have proxy. Reviewboard is installed on local computer, it's url is http://localhost:8090/reviewboard. -- Get the Review Board Power Pack at http://www.reviewboard.org/powerpack/ --- Sign up for Review Board hosting at RBCommons: https://rbcommons.com/ --- Happy user? Let us know at http://www.reviewboard.org/users/ --- You received this message because you are subscribed to the Google Groups "reviewboard" group. To unsubscribe from this group and stop receiving emails from it, send an email to reviewboard+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: Reviewboard 2.0.6 hungs on new request with non latin symbols
A lot of people use Firefox with Review Board. Are there any extensions you’re using? Is there an HTTP proxy configured? Perhaps that’s interfering. - Christian -- Christian Hammond - christ...@beanbaginc.com Review Board - http://www.reviewboard.org Beanbag, Inc. - http://www.beanbaginc.com On September 2, 2014 at 10:03:17 PM, kovalyov...@gmail.com (kovalyov...@gmail.com) wrote: I opened firebug network panel before making new request. There are no errors in firebug. I founded workaround - using Google Chrome instead of Firefox. Maybe it is firefox problem? -- Get the Review Board Power Pack at http://www.reviewboard.org/powerpack/ --- Sign up for Review Board hosting at RBCommons: https://rbcommons.com/ --- Happy user? Let us know at http://www.reviewboard.org/users/ --- You received this message because you are subscribed to the Google Groups "reviewboard" group. To unsubscribe from this group and stop receiving emails from it, send an email to reviewboard+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout. -- Get the Review Board Power Pack at http://www.reviewboard.org/powerpack/ --- Sign up for Review Board hosting at RBCommons: https://rbcommons.com/ --- Happy user? Let us know at http://www.reviewboard.org/users/ --- You received this message because you are subscribed to the Google Groups "reviewboard" group. To unsubscribe from this group and stop receiving emails from it, send an email to reviewboard+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: Reviewboard 2.0.6 hungs on new request with non latin symbols
I opened firebug network panel before making new request. There are no errors in firebug. I founded workaround - using Google Chrome instead of Firefox. Maybe it is firefox problem? -- Get the Review Board Power Pack at http://www.reviewboard.org/powerpack/ --- Sign up for Review Board hosting at RBCommons: https://rbcommons.com/ --- Happy user? Let us know at http://www.reviewboard.org/users/ --- You received this message because you are subscribed to the Google Groups "reviewboard" group. To unsubscribe from this group and stop receiving emails from it, send an email to reviewboard+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: Reviewboard 2.0.6 hungs on new request with non latin symbols
If the last API request was to fetch the commits list, then it's not even trying to send the patch to the server. Trying to set up a repro case with your particular files won't help because it's a problem in the javascript before it ever actually deals with the file. Are you sure there are no errors in firebug? Try opening the panel before doing anything on the page. -David On Mon, Sep 1, 2014 at 11:07 PM, wrote: > I don't see any error In network panel of firebug. > Last row is > http://localhost:8090/reviewboard/api/repositories/2/commits/?branch=Trunk&api_format=json > and i'ts status is ok. > > Can You reproduce this error with my test case (using test.py and > test.py.patch)? > > -- > Get the Review Board Power Pack at http://www.reviewboard.org/powerpack/ > --- > Sign up for Review Board hosting at RBCommons: https://rbcommons.com/ > --- > Happy user? Let us know at http://www.reviewboard.org/users/ > --- > You received this message because you are subscribed to the Google Groups > "reviewboard" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to reviewboard+unsubscr...@googlegroups.com. > For more options, visit https://groups.google.com/d/optout. > -- Get the Review Board Power Pack at http://www.reviewboard.org/powerpack/ --- Sign up for Review Board hosting at RBCommons: https://rbcommons.com/ --- Happy user? Let us know at http://www.reviewboard.org/users/ --- You received this message because you are subscribed to the Google Groups "reviewboard" group. To unsubscribe from this group and stop receiving emails from it, send an email to reviewboard+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: Reviewboard 2.0.6 hungs on new request with non latin symbols
I don't see any error In network panel of firebug. Last row is http://localhost:8090/reviewboard/api/repositories/2/commits/?branch=Trunk&api_format=json and i'ts status is ok. Can You reproduce this error with my test case (using test.py and test.py.patch)? -- Get the Review Board Power Pack at http://www.reviewboard.org/powerpack/ --- Sign up for Review Board hosting at RBCommons: https://rbcommons.com/ --- Happy user? Let us know at http://www.reviewboard.org/users/ --- You received this message because you are subscribed to the Google Groups "reviewboard" group. To unsubscribe from this group and stop receiving emails from it, send an email to reviewboard+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: Reviewboard 2.0.6 hungs on new request with non latin symbols
In Firebug, please open up the network panel and then reproduce your issue. You should see a request to the server that's failing (a 4XX or 5XX error). Open up that request, and please send us the contents of the "Headers" and "Response" tabs. -David On Mon, Sep 1, 2014 at 2:16 AM, wrote: > Below simple test on this error. > > 1) Add test.py from attachement to svn working folder > 2) Select "New Request" from reviewboard > 3) Select test.py.patch as diff file in reviewboard > 4) Specify svn base direcory > 5) Click Ok button > > After these actions I see "Loading" message. This message don't go away. > > -- > Get the Review Board Power Pack at http://www.reviewboard.org/powerpack/ > --- > Sign up for Review Board hosting at RBCommons: https://rbcommons.com/ > --- > Happy user? Let us know at http://www.reviewboard.org/users/ > --- > You received this message because you are subscribed to the Google Groups > "reviewboard" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to reviewboard+unsubscr...@googlegroups.com. > For more options, visit https://groups.google.com/d/optout. > -- Get the Review Board Power Pack at http://www.reviewboard.org/powerpack/ --- Sign up for Review Board hosting at RBCommons: https://rbcommons.com/ --- Happy user? Let us know at http://www.reviewboard.org/users/ --- You received this message because you are subscribed to the Google Groups "reviewboard" group. To unsubscribe from this group and stop receiving emails from it, send an email to reviewboard+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: Reviewboard 2.0.6 hungs on new request with non latin symbols
Below simple test on this error. 1) Add test.py from attachement to svn working folder 2) Select "New Request" from reviewboard 3) Select test.py.patch as diff file in reviewboard 4) Specify svn base direcory 5) Click Ok button After these actions I see "Loading" message. This message don't go away. -- Get the Review Board Power Pack at http://www.reviewboard.org/powerpack/ --- Sign up for Review Board hosting at RBCommons: https://rbcommons.com/ --- Happy user? Let us know at http://www.reviewboard.org/users/ --- You received this message because you are subscribed to the Google Groups "reviewboard" group. To unsubscribe from this group and stop receiving emails from it, send an email to reviewboard+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout. # -*- coding: utf-8 -*- """ Тест """ from __future__ import unicode_literals Index: test.py === --- test.py (revision 0) +++ test.py (working copy) @@ -0,0 +1,9 @@ +# -*- coding: utf-8 -*- + +""" + +Тест + +""" + +from __future__ import unicode_literals
Re: Reviewboard 2.0.6 hungs on new request with non latin symbols
There is no errors in reviewboard log. Can You tell me how to check errors in browser debug console? I am using firefox. I have run Firebug, but don't see any errors. I see message "Loading..." at browser, and this message don't go away. -- Get the Review Board Power Pack at http://www.reviewboard.org/powerpack/ --- Sign up for Review Board hosting at RBCommons: https://rbcommons.com/ --- Happy user? Let us know at http://www.reviewboard.org/users/ --- You received this message because you are subscribed to the Google Groups "reviewboard" group. To unsubscribe from this group and stop receiving emails from it, send an email to reviewboard+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: Reviewboard 2.0.6 hungs on new request with non latin symbols
Are there any errors in the Review Board log or the browser debug console? -David On Thu, Aug 28, 2014 at 6:09 AM, wrote: > I am using Reviewboard 2.0.6 with Postgres database in utf-8. > When I making new pre-commit request on svn-based repository with > diff-file containing non-latin symbols, reviewboard hungs. > If diff file contains only latin symbols, it works ok. > > Is it possible to fix this bug? > > -- > Get the Review Board Power Pack at http://www.reviewboard.org/powerpack/ > --- > Sign up for Review Board hosting at RBCommons: https://rbcommons.com/ > --- > Happy user? Let us know at http://www.reviewboard.org/users/ > --- > You received this message because you are subscribed to the Google Groups > "reviewboard" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to reviewboard+unsubscr...@googlegroups.com. > For more options, visit https://groups.google.com/d/optout. > -- Get the Review Board Power Pack at http://www.reviewboard.org/powerpack/ --- Sign up for Review Board hosting at RBCommons: https://rbcommons.com/ --- Happy user? Let us know at http://www.reviewboard.org/users/ --- You received this message because you are subscribed to the Google Groups "reviewboard" group. To unsubscribe from this group and stop receiving emails from it, send an email to reviewboard+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
