View Issue Details
| ID | Project | Category | View Status | Date Submitted | Last Update |
|---|---|---|---|---|---|
| 0005724 | Taler | twister | public | 2019-05-15 14:58 | 2019-12-20 18:12 |
| Reporter | Marcello Stanisci | Assigned To | Marcello Stanisci | Resolution deadline | |
| Implementer | Reviewer | ||||
| Release signed by | Deployed by | ||||
| Verifier | Communicator | ||||
| Data protection authority notification required | No | ||||
| CSIRT notification required | No | ||||
| Estimated effort (hours) | Actual effort (hours) | 0 | |||
| Fast-track | No | ||||
| Work package | none | ||||
| Priority | normal | Severity | minor | Reproducibility | have not tried |
| Status | closed | Resolution | fixed | ||
| Product Version | git (master) | ||||
| Target Version | 0.6 | Fixed in Version | 0.6 | ||
| Summary | 0005724: code repetition and conflicting parameters. | ||||
| Description | The Twister has some code repetition where both the POST and the PUT cases set the CURLOPT_INFILESIZE option. And more, this option sounds like conflicting -- or doing "the same thing" -- as CURLOPT_POSTFIELDSIZE. We should either not repeat the code, or cut entirely off the CURLOPT_INFILESIZE paths. | ||||
| Tags | No tags attached. | ||||
| As per manpage, CURLOPT_POSTFIELDSIZE should be used when the payload size should NOT be calculated by libcurl; for example, when binary files are being sent. That's due to the fact that libcurl uses strlen() to calculate that size. | |
| CURLOPT_INFILESIZE, OTOH, sounds like to be used when uploading *files*, that we also never do with the Twister. So in short, the two options look not important for the Twister. Agree? | |
| Well, testing confirms that POSTs want CURLOPT_POSTFIELDSIZE and PUTs want CURLOPT_INFILESIZE. The real optimization was to set all the options (including read/write callbacks) but the method on the easy handle *before* even checking which method we are going to perform, and set the method as a last step. This is not breaking, since if a method doesn't need a callback/option, it will ignore it. The gain is that all the common options set by the *three* methods (GET/POST/PUT) are mentioned in *one* place in the code, instead of three. Tested against the Twister testcase itself, and on exchange's and merchant's twisted tests. | |
| Date Modified | Username | Field | Change |
|---|---|---|---|
| 2019-05-15 14:58 | Marcello Stanisci | New Issue | |
| 2019-05-15 14:58 | Marcello Stanisci | Status | new => assigned |
| 2019-05-15 14:58 | Marcello Stanisci | Assigned To | => Marcello Stanisci |
| 2019-06-03 18:27 | Marcello Stanisci | Note Added: 0014502 | |
| 2019-06-03 18:33 | Marcello Stanisci | Note Added: 0014503 | |
| 2019-06-03 19:17 | Marcello Stanisci | Note Added: 0014504 | |
| 2019-06-03 19:18 | Marcello Stanisci | Note Edited: 0014504 | |
| 2019-06-03 19:18 | Marcello Stanisci | Note Edited: 0014504 | |
| 2019-06-03 19:22 | Marcello Stanisci | Note Edited: 0014504 | |
| 2019-06-03 19:22 | Marcello Stanisci | Note Edited: 0014504 | |
| 2019-06-03 20:16 | Marcello Stanisci | Status | assigned => informing |
| 2019-06-03 20:16 | Marcello Stanisci | Resolution | open => fixed |
| 2019-06-05 19:28 | Christian Grothoff | Product Version | => git (master) |
| 2019-06-05 19:28 | Christian Grothoff | Fixed in Version | => 0.6 |
| 2019-06-05 19:28 | Christian Grothoff | Target Version | => 0.6 |
| 2019-12-20 18:12 | Christian Grothoff | Status | informing => closed |