On Wed, Sep 25, 2024 at 15:44 Yihong Yu ***@***.***> wrote: ***@***.**** commented on this pull request. ------------------------------ In tests/test_rule_generator.py <#48 (comment)>: > @@ -2162,7 +2162,7 @@ def test_generate_general_rule_16(): q0_rule, q1_rule = unify_variable_names(rule['pattern'], rule['rewrite']) assert q0_rule== "SELECT <x1>, <x2>, <x3>, <x4>, <x5>, <x6> FROM <x7> WHERE <x2> IN (SELECT <x2> FROM <x7> WHERE <x6> = <x8> AND <x3> = <x9>) ORDER BY <x2>, <x3>" - assert q1_rule == "SELECT <x10>.<x1>, <x10>.<x2>, <x10>.<x3>, <x10>.<x4>, <x10>.<x5>, <x10>.<x6> FROM <x10> JOIN <x11> ON <x11>.<x2> = <x10>.<x2> WHERE <x11>.<x6> = <x8> AND <x11>.<x3> = <x9> ORDER BY <x10>.<x2>, <x10>.<x3>" + assert q1_rule == "SELECT <x7>.<x1>, <x7>.<x2>, <x7>.<x3>, <x7>.<x4>, <x7>.<x5>, <x7>.<x6> FROM <x7> JOIN <x10> ON <x10>.<x2> = <x7>.<x2> WHERE <x10>.<x6> = <x8> AND <x10>.<x3> = <x9> ORDER BY <x7>.<x2>, <x7>.<x3>" In the example query, we have FROM historicoestatusrequisicion hist1 JOIN historicoestatusrequisicion hist2. In our current logic, hist2 will be represented as <x10> to indicate that a JOIN is being performed on another table (even though hist1 and hist2 are the same table). As we discussed in July, it's still difficult for QueryBooster to determine if two tables with different alias are the same without refactoring the code. This is only a temporary fix and the user still needs to manually replace to make the rule meaningful in a pairwise comparison 😢 — Reply to this email directly, view it on GitHub <#48 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AB542BXSST5ULEZHUHAUMFTZYM4FFAVCNFSM6AAAAABO3QLXF2VHI2DSMVQWIX3LMV43YUDVNRWFEZLROVSXG5CSMV3GSZLXHMZDGMRZGY2TONRZGM> . You are receiving this because you commented.Message ID: ***@***.***>