- Jun 09, 2023
-
-
id can have `-` in it, and `\w+` does not match them. We change for `.+`
-
- May 23, 2023
-
-
- May 02, 2023
-
-
Information regarding stocks' general assembly was wrongly considered as an account
-
- Apr 08, 2023
-
-
- Mar 01, 2023
-
-
Url an xpath have changed for card fetching in professionnal account space. Cards were not scraped anymore without rasing any exception.
-
- Feb 11, 2023
-
-
Romain Bignon authored
-
- Jan 26, 2023
-
-
Filter error because we did not ignore non-investment items that do not have quantity, unitvalue, etc.
-
Remove BrowserForbidden, in BankStatementPage.check_error() raised in iter_subscription in this specific case. This Exception occured when account owner has not asked for enabling e-document feature.
-
- Jan 10, 2023
-
-
Societe Generale becomes SG
-
Error 'fiabilisation_coordonnees' occurs when the user needs to validate its contact information on its personal space
-
- Dec 09, 2022
-
-
jlouvel authored
New types of pro loans are fetched since a few days. A few particular cases were not covered. Some loans must be skipped since they contain almost no information (Crédit Bail Mobilier or Location Option d'Achat for example). Pro loan page can sometimes return an unavailable status even on the website, handling of that message have been added to prevent the for loop fetching the accounts stopping. Also added a default NotAvailable for an element of the label that is called "informationFacultative" and, as its name says, this value of this key is not always provided in the JSON. Finally, added a condition to skip some PER and PERCO. These are accounts that have no information on the website except a label and a contract number. Clicking on them on the website leads us to an error page.
-
jlouvel authored
When fetching short-term credits, we can have "Autorisation de découvert (Convention de Trésorerie Courante)" accounts. Those accounts only have an iban and a "maximum amount". These are not regular loans so we skip them. Also added defaults for these short and medium-term credits, data for attributes last_payment_amount and last_payment_date (or next_) can be null on the website for some loans.
-
sfartit authored
Some pages that contains pro loans (short term, and medium/long term) were not scraped before. Also a page that contains life insurances and per accounts was missing. This MR should handle them.
-
- Jul 04, 2022
-
-
Since most such requests work and only some fail with this reason, we suppose its a temporary unavailability and add a retry.
-
- Jun 27, 2022
-
-
ntome authored
-
- Jun 13, 2022
-
-
Management of unhandled errors in the spge part of the module. Sometimes in the spge part of the module, we get the value 'nok' for the key ‘state’ in the json for different 'reasons', which can be ENCADREMENT_KYC_PREAVIS, ENCADREMENT_KYC_POST_PREAVIS, INSCRIP_OBL, in this case we raise ActionNeeded because the user has to fill in some information on the bank website. We also manage the ‘err_is’ and ‘err_tech’ cases in which we raise BrowserUnavailabe as in the browser of the main module. Sometimes the key 'reason' has the value 'FIABILISATION_TS'. To manage this case in the main module there is a function named check_skippable_action_needed. However, in some cases the function does not seem to be sufficient to pass FIABILISATION_TS state. In this case, we have the key 'chgtnivauth' in the returned json. During login, this key is also present when an SCA is required. It is likely that, in some cases, an SCA is also required to pass the 'FIABILISATION_TS' state. So, in the main module, I add the possibility to trigger an SCA if the json obtained in the check_skippable_action_needed function contains the chgtnivauth key. We do not have a PSU to test the solution regarding FIABILISATION_TS.
-
- May 30, 2022
-
-
When reaching the subscriptions, the user sometimes reaches errors; those are similar to some errors reached in the accounts pages. We thus reuse a similar method to handle reached errors. Unencountered errors will have to be added manually later. Not tested : sca prevents us to really test, and our sgpe creds are down.
-
- Apr 28, 2022
-
- Mar 16, 2022
-
-
update the Bill URLs, iter_subscriptions and iter_documents.
-
- Feb 28, 2022
-
-
-
-
ntome authored
basestring, range, long, url*, parse*, unicode, html_unescape also, remove some obsolete sys.version checks
-
- Feb 18, 2022
-
-
ntome authored
-
- Jan 13, 2022
-
-
Florent Viard authored
Fixes a regression: iter_recipients func definition was modified in societegenerale but not inside the sgpe child.
-
- Nov 24, 2021
-
-
If the URL contained in the storage is the recipients URL, the website returns a JSON telling us we're not logged. We're now using this JSON to set a logged property on the concerned pages. Also changed the "on_load" methods for a manual check_error method.
-
- Oct 13, 2021
-
-
- Sep 18, 2021
-
-
hydrargyrum authored
-
- Sep 13, 2021
-
-
Sometimes, we get several cards even though we asked for one. They have the same number but some might not have a date. Instead of getting the first card, we get the first one with a date.
-
- Aug 13, 2021
-
-
- self.response.json()['donnees'] = null so there is no number to get (get_number()) therefore no comings
-
- Jun 30, 2021
-
-
- May 10, 2021
-
-
- Apr 26, 2021
-
-
- Apr 04, 2021
-
-
- Feb 26, 2021
-
-
When reloading state and requesting the last page visited, we are usally logged out. If that page was EasyTransferPage, in that case, it is now a JSON, instead of the regular HTML when already logged. This is now handled with a special logged property.
-
- Feb 12, 2021
-
-
When reloading state and requesting the last page visited, when that page is one of the one in json_pages.py, we are usally logged out. This is handled in AccountJsonPage with a special Logged property. But it needs to be as well in the other pages. We do that by introducing a LoggedDetectionMixin, inherited by a SGPEJsonPage, common to all subsequent pages.
-
To better avoid problems when changing things not common to SGEntreprise and SGProfessional, inheritance scheme is changed. A common SGPEBrowser still exists. It takes in common login steps from SocieteGeneraleLogin, and common Pro/Ent navigation. But SGProfessionalBrowser and SGEnterpriseBrowser are now each inheriting separately from SGPEBrowser. It bears importance for the few differences in 2FA handling for Ent, where it is systematic, compared to Pro. Also, dead code is deleted (browser and pages' side). And a few commenting is added.
-
Initially only the case of "PLAFOND_SIGNATURE_DEPASSE" had to be added. But after discussions, the need to fix other cases emerged.
-
-