AutoED reports

AutoED reports are files that give an overview of the data processing statistics. These files are generated automatically as the data is being processed. The reports are saved in the report directory (i.e., autoed_report). For each watched directory, the report directory appears at the same level as the data root directory (e.g., the ED directory) and the directory for the processed results. Currently, reports are available as a single HTML file or two textual (txt) files. If you look inside autoed_report, you might find something like this

report_data
report.html
report_sorted.txt
report.txt
server

The report_data is a directory where we keep all the relevant data necessary to generate an HTML report (e.g., beam position figures, xia2 reports, etc). The report.html is not a static HTML page. Therefore, you need to start a local HTTP server to view it. Run the server script

server

This command will start a server on a local port 8000. You can see the generated report in your browser at http://localhost:8000/report.html. To kill the HTTP server when you finish viewing, run

server stop

Your browser must run on the same machine where the report is located. So, if you are on a Diamond filesystem, you must access a remote desktop using NoMachine. Another option is downloading the AutoED report directory to your local machine and running the server there.

If you want to view multiple reports:

  • You can always start another server at a different port (e.g., port 8001) with

    server 8001

    and then go to http://localhost:8001/report.html.

  • You can run a server on the default 8000 port, which will overwrite your previous report. You might need to refresh your browser to see the new report (e.g., for Firefox on Ubuntu, the refresh is <ctrl-F5>).

If viewing the HTML report is too complicated, you can view the textual report. There are two plain text files with processing statistics. report.txt will give you a chronological list of datasets as they are processed (stored in the database). On the other hand, report_sorted.txt will provide you with a list of only those datasets that were processed successfully and ordered according to the percentage of indexed points.

The textual reports are concise because they do not provide the results from all the pipelines. Instead, they pick a pipeline with the highest percentage of indexed spots and only list that one.