aboutsummaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2025-05-19Fix generic bugsFrederick Muriuki Muriithi
2025-05-19Import missing symbols.Frederick Muriuki Muriithi
2025-05-19Provide extra metadata to the job.Frederick Muriuki Muriithi
2025-05-19Provide missing URI to the MariaDB database to the script.Frederick Muriuki Muriithi
2025-05-19Cross-reference phenotypes, publications, and data.Frederick Muriuki Muriithi
2025-05-19Retrieve publication first.Frederick Muriuki Muriithi
2025-05-19Simplify handling of publications: Assume a maximum of one.Frederick Muriuki Muriithi
Adding publication details in the R/qtl2 bundle might not be an easy thing, so for now, assume all the phenotypes uploaded in a particular session are all published in one publication.
2025-05-19Add placeholder stub to be implemented later.Frederick Muriuki Muriithi
2025-05-19Add some notes for later.Frederick Muriuki Muriithi
2025-05-19Save numeric/computational data for phenotypes.Frederick Muriuki Muriithi
2025-05-19Retrieve samples' details from the database.Frederick Muriuki Muriithi
2025-05-19Add function to save the numerical data for phenotypes.Frederick Muriuki Muriithi
2025-05-19Use correct files for basic phenotypes data. Add notes to self.Frederick Muriuki Muriithi
The basic phenotype data will be in files under the key "gn-metadata" -> "pheno". We support the possibility of having some descriptions in the "phenocovar" files, since that can happen also.
2025-05-19Rework order of execution of potential steps.Frederick Muriuki Muriithi
2025-05-19Cleanup function call.Frederick Muriuki Muriithi
2025-05-19Fetch next available PublishXRef.DataId value for the phenotypes.Frederick Muriuki Muriithi
2025-05-19Save the basic phenotypes data into the database.Frederick Muriuki Muriithi
2025-05-19Read headers from a CSV file, whether transposed or not.Frederick Muriuki Muriithi
2025-05-19Pass Connection rather than Cursor: Transaction is maintained.Frederick Muriuki Muriithi
The idea is that all the data is saved to the database in a single transaction, rather than in bits that could lead to data inconsistencies. As it were, simply passing the connection object, and letting each function create its own cursor will still allow the transaction to be maintained and will not necessitate the refactor of multiple already existing functions.
2025-05-19Initialise function to save publicationsFrederick Muriuki Muriithi
Do a rudimentary save of the publications: this is incomplete and probably very buggy.
2025-05-19Setup log format for the application.Frederick Muriuki Muriithi
2025-05-12Wireframe saving the data into the database.Frederick Muriuki Muriithi
2025-05-12Print out progress to STDOUT.Frederick Muriuki Muriithi
2025-05-12Use builtin functions to compute log level.Frederick Muriuki Muriithi
2025-05-12Pass in the logger, rather than the name of the logger.Frederick Muriuki Muriithi
2025-05-12Handle exceptions within the `with` to prevent silent failureFrederick Muriuki Muriithi
One, or both of the context managers is "swallowing" exceptions, leading to silent failures. This change manually handles the exceptions within the context manager to avoid such silent failures.
2025-05-12Break "merge file" race-conditionFrederick Muriuki Muriithi
Multiple requests/threads were attempting to merge the same file, leading to race conditions, and sometimes, rarely, it would lead to actual data corruption.
2025-05-12Allow lowercase versions of the log-level choices.Frederick Muriuki Muriithi
2025-05-12BugFix: Don't scroll to top of page on clicking "Browse" buttons.Frederick Muriuki Muriithi
This fixes the behaviour where clicking the "Browse" buttons on any of the resumable elements, the viewport would scroll to the top of the page.
2025-05-09Use more of the screen width for content on smaller screens.Frederick Muriuki Muriithi
Display only the links and main content on screens smaller than 20 CSS inches — the idea is that laptop screens and smaller will only display the links and the main content, leaving the sidebar content hidden.
2025-05-05Init setup and script for async job to load phenotypes into databaseFrederick Muriuki Muriithi
2025-05-05Save/Create new phenotypes in the database.Frederick Muriuki Muriithi
2025-05-05Provide endpoint for loading data after QC is successful.Frederick Muriuki Muriithi
2025-05-05Require that the publication is present.Frederick Muriuki Muriithi
2025-05-05Move previews below each item rather than being in the sidebar.Frederick Muriuki Muriithi
2025-05-05Include the publication when provided.Frederick Muriuki Muriithi
2025-05-05Allow plain text files too.Frederick Muriuki Muriithi
2025-05-02Mark directories as packages to make modules unique.Frederick Muriuki Muriithi
2025-05-02PoC: Use independent module-level loggers with applicationFrederick Muriuki Muriithi
Using flask.current_app.logger for logging in modules leads to the modules requiring an app context, even when the module itself could be used outside of the application. This commit provides a proof-of-concept (really, me finally putting in the effort to figure out logging) for how we could allow independent loggers at the module level, that can then be hooked up to the main logging in the system in question.
2025-05-02Format abstracts a little better.Frederick Muriuki Muriithi
2025-04-28Provide correct redirection uri, and data.Frederick Muriuki Muriithi
2025-04-28Encode data to UTF-8 to prevent persistence errors.Frederick Muriuki Muriithi
2025-04-28Search for radio was slow for some reason.Frederick Muriuki Muriithi
2025-04-28Allow selection from existing publications.Frederick Muriuki Muriithi
2025-04-28Do not auto-update publication details on conflictFrederick Muriuki Muriithi
Auto-update of a publication's details on an "insert conflict" would lead to a bug where publications could be overwritten with the wrong information. Rather than auto-update, explicitly indicate the error, and let the user manually edit the record to avoid data inconsistency.
2025-04-28Implement creating publications.Frederick Muriuki Muriithi
2025-04-28Add logging.Frederick Muriuki Muriithi
Without this logging statement, the chunk merging fails with an error that it cannot find one of the chunks when attempting to delete it. That error disappears as soon as I attempt to log out the error and debug.
2025-04-28Enable autofill of information from PubMed.Frederick Muriuki Muriithi
2025-04-28Move functions to fetch pubmed data out into a separate js file.Frederick Muriuki Muriithi
2025-04-28Move generic utility functions to the utils.js file.Frederick Muriuki Muriithi