pandas/merge-on-both-indexes
| Rule | merge-on-both-indexes |
|---|---|
| Pack | pandas |
| Severity | info |
| Fix | unsafe, with xpd check --fix --unsafe-fixes |
| Library versions | every version |
A merge on both indexes is a join, but the merge call hides that rows are paired by index label. Call join with the same how instead.
both = left.merge(right, left_index=True, right_index=True)both = left.join(right, how="inner")What the rule reports
Section titled “What the rule reports”The rule reports left.merge(right, left_index=True, right_index=True) and the pd.merge form. The merge gave a frame equal to left.join(right, how="inner"), and the two were also equal for how="left", "right" and "outer" and with validate="1:1". The fix is unsafe, because join raises ValueError when the two frames share a column name. merge adds the suffixes _x and _y to such columns.
How to fix it
Section titled “How to fix it”xpd check --fix --unsafe-fixes applies the fixes of this rule. Check the result, because an unsafe fix can change what the code does.
The unsafe fix writes left.join(right, how="inner"). None for a merge with another keyword than how, validate and sort.
Message
Section titled “Message”xpd prints this message with the issue. A name that starts with $ stands for code from your file.
A merge on both indexes is a join. Write $A.join(..., how=...), which says that the rows are paired by index label. merge keeps the shared labels unless how= says otherwise, which is how="inner" in join.Browse the other rules of the pandas pack.