The key field starts [' and ends '] and this should be stripped out as it is never used.

Currently only allows a single user ID, Google account, and when the first run for the first time then prompt for the user and the permission. There is no way to switch to a different user ID.

it would be better if you could specify option in the command line for which Google account you want to run under for this run. it would be even better if you could pass a filename of a text file that contains a list of Google account, and then run the command once each for each of those Google accounts. of course during the first run for a new Google account in will have to prompt for permission, but after that it should save their credentials.

When the token is empty it should not insert a hyphen before the final names

Under some circumstances, the output xlsx file has columns named "key1" and "key2" etc. This prevents using Excel pivot DAX as key* is a reserved word. Changing to something else like "keys-1"

Both scripts, the GSC and the GA one, use the same get services function based on the sample code from Google.

Split this function out into a separate .py file that can be imported as a module.

Therefore upgrades to this function need only to be done once.

At the moment it is hardcoded to "web", give a command line choice for video or image.

implement a -d --debug switch that takes a number such as 0, 1, or 3 to change behaviour to more or less debugging

The NewDownload.py script which downloads from Google Search Console needs to output an "options" sheet that details the options of that report.

Same things as the GA script is already done, just need to transfer that functionality.

Hide the excessive debug output and do a progress bar like the other script. Or keep the debug output but hide it behind a "--debug" flag?

When opening the xlsx file I find that the number columns are all "Number stored as text". Although it is easy to fix it would be better if the code did this as it has to be done every time.

Make the automatically generated filenames of report .xlsx more informative. Include the options that are in the command line parameters become part of the filename.

instead of:

analytics-2020-02-25-13-16-41.xlsx

automatically create a filename with more useful information

analytics-generated-2020-02-25-13-16-41-date-range-2020-01-01-2020-01-31-days-31-filtered-page-views-gt-2-dimension-pagePath-metric-adsenseRevenue.xlsx

Probably should make it an option as not every time will want all those details.

`GACombined2.py -m adsenseRevenue,ga:adsenseAdUnitsViewed yesterday today

gives an error

ValueError: Length mismatch: Expected axis has 3 elements, new values have 2 elements

This applies to the Google Analytics output. Even when the property/view has been set to https the script still puts http:// into the output file.

Every time I use the GSC output I create some additional analysis columns, it would save a little time if that was automated by the script:

• For reports showing "siteUrl" add "root domain" column - "example.com" instead of "https://www.example.com"
• For reports showing a "key" which is a url show the stripped url - "example.com/path/filename.html" instead of "https://www.example.com/path/filename.html"
• For reports showing a "query" add "number of words" - "this is a query" = 4

These should be additional columns.

It is easy to forget what options were used during the run when looking at an output file later.

Create a second sheet in the xlsx with some text fields that just show what arguments were used.

Only sometimes. Maybe if a site doesn't have required fields? Probably ok just to skip that item

GACombined2.py -d ga:source -m ga:sessions,ga:goalCompletionsAll

Traceback (most recent call last):
File "\dell\business\GitHub\Google-analytics-and-search-console\GACombined2.py", line 187, in
combinedDF[splitMetrics] = combinedDF[splitMetrics].apply(pd.to_numeric)
File "C:\Users\raymo\AppData\Local\Programs\Python\Python38\lib\site-packages\pandas\core\frame.py", line 2912, in getitem
indexer = self.loc._get_listlike_indexer(key, axis=1, raise_missing=True)[1]
File "C:\Users\raymo\AppData\Local\Programs\Python\Python38\lib\site-packages\pandas\core\indexing.py", line 1254, in _get_listlike_indexer
self._validate_read_indexer(keyarr, indexer, axis, raise_missing=raise_missing)
File "C:\Users\raymo\AppData\Local\Programs\Python\Python38\lib\site-packages\pandas\core\indexing.py", line 1298, in _validate_read_indexer
raise KeyError(f"None of [{key}] are in the [{axis_name}]")
KeyError: "None of [Index(['ga:sessions', 'ga:goalCompletionsAll'], dtype='object')] are in the [columns]"

-t n / --test n where n defaults to 3

To speed up development have a switch that only does a test of the first n number of things.

So for every loop such as loop through profiles or loop through accounts exit the loop after n

Such as page and query. So should output one row per page/query combination.

Will help to speed things up slightly by omitting excessive data rows

Suggest the filename is something like "search-console-2019-04-01-2019-04-07-created-2019-04-01-13-45-67.xlsx"

Should specify which version of python and packages are required. Preferably just a list of PIP commands that can be cut-and-pasted to set up.