I tried doing this using this example listed here https://towardsdatascience.com/introduction-to-git-data-extraction-and-analysis-in-python-e7e2bf9b4606
But the count obtained from the above method didn't match with the count displayed in 'Commit Activity' section for repository 'activemq' https://github.com/apache/activemq/graphs/commit-activity.
This is probably because the merge commits are not included in Github insights.
For example apache/activemq Insights lists 2 commits on 15th feb 2022.
But if I run the the commit search for the main branch on 15th I get 3 commits (I am using a graphQL for the search).
{
repository(name: "activemq", owner: "apache") {
ref(qualifiedName: "main") {
target {
... on Commit {
history(since: "2022-02-15T00:00:00+00:00", until: "2022-02-16T00:00:00+00:00") {
totalCount
nodes{
authoredDate
message
}
}
}
}
}
}
}
The result of the above query displays three commits on the 15th. But the second commit is a merge commit and hence not included in the insights.
{
"data": {
"repository": {
"ref": {
"target": {
"history": {
"totalCount": 3,
"nodes": [
{
"authoredDate": "2022-02-15T06:35:30Z",
"message": "Fix version in SECURITY.md"
},
{
"authoredDate": "2022-02-15T04:38:56Z",
"message": "Merge pull request #767 from coheigea/coheigea/SECURITY.md\n\nAdding SECURITY.md for ActiveMQ"
},
{
"authoredDate": "2022-02-15T04:34:35Z",
"message": "Adding SECURITY.md for ActiveMQ"
}
]
}
}
}
}
}
}