Hacktoberfest 2026: le issue che i maintainer hanno segnato per ottobre, aperte e adatte ai principianti. Sfoglia le issue Hacktoberfest

pgtap/mincut/compare.test.sql test data is flawed

Aperta
  • Qualcosa non va?
#2,583 3 commenti 1 reazione 0 assegnatari Vedi su GitHub

I maintainer di solito rispondono entro 2 giorni

Nessuno ha ancora preso questa issue.

  • #3104 di @prashsti29 — chiusa senza merge

Valutazione

Difficoltà
3/5
Tempo stimato
1-2 giorni
Idoneità per principianti
65/100
Tipo di issue
Bug
Chiarezza
Specificata chiaramente
Stato di attività
Tranquilla
Stack tecnologico
cpp, postgresql, sql
Ambito
databases, testing

Direzione di ricerca

The issue is in the test file pgtap/mincut/compare.test.sql. First, examine the flawed test data for tests 5 and 6, understanding the graph structure and the Stoer-Wagner algorithm requirements. Review the pgr_stoerWagner function's implementation and the Boost library version impact. Create a valid test graph with non-negative costs and undirected edges, then update the test SQL with correct expected results. Run the TAP tests to verify the fix.

Scritto dal modello di indicizzazione a partire dal testo della issue.

Descrizione

bug mincut pgtap

Problem
The two functionality tests in pgtap/mincut/compare.test.sql use graphs that are either vacuous or invalid, and the expected return values are not guaranteed.

This test case would benefit from being rewritten to use a graph designed for min-cut testing, but I am raising this issue because changes to Boost can cause it to fail.

To Reproduce
Compiling the test against Boost 1.80 or later (which uses a different implementation of Stoer-Wagner than earlier versions) causes failures.

Test 5

The graph for unit test "5: Mincut of edge 17" uses a graph with two nodes and one edge. Evaluating the inner query of the PREPARE stoerWagner6 statement (the one being passed to pgr_stoerWagner()) gives the following:

 id | source | target | cost | reverse_cost 
----+--------+--------+------+--------------
 17 |     14 |     15 |    1 |            1
(1 row)

Obviously this is the exact same graph as generated by the simpler stoerWagner5 subquery; in both cases the mincut is trivially to cut the only existing edge, so the comparison of stoerWagner5 and stoerWagner6 tests only that pgr_stoerWagner generates the same output given the same input, which was probably not what was intended.

Test 6

The subquery being passed to pgr_stoerWagner7 for the test "6: Compare the mincut of subgraph with actual result" results in the following graph:

 id | source | target | cost | reverse_cost 
----+--------+--------+------+--------------
  1 |      1 |      2 |    1 |            1
  2 |      2 |      3 |   -1 |            1
  3 |      3 |      4 |   -1 |            1
  4 |      2 |      5 |    1 |            1
  5 |      3 |      6 |    1 |           -1
  6 |      7 |      8 |    1 |            1
  7 |      8 |      5 |    1 |            1
  8 |      5 |      6 |    1 |            1
  9 |      6 |      9 |    1 |            1
 10 |      5 |     10 |    1 |            1
 11 |      6 |     11 |    1 |           -1
 12 |     10 |     11 |    1 |           -1
 13 |     11 |     12 |    1 |           -1
 14 |     10 |     13 |    1 |            1
 15 |      9 |     12 |    1 |            1
 16 |      4 |      9 |    1 |            1
(16 rows)

And expects that the mincut result is to cut the single edge with id 1.

There are three flaws with this test:

  1. The Stoer-Wagner algorithm requires all edges have non-negative costs, but edges 2 and 3 violate this.
  2. The Stoer-Wagner algorithm requires an undirected graph, but edges 2, 3, 5, 11, 12, and 13 have different costs in different directions.
  3. There are four distinct edges of equal weight that, each on its own, can be removed to partition the graph: edges 1, 6, 7, and 14. (Notice that nodes 1, 7, and 13 only have a single connection to the graph, and node 8 lies between node 7 and the rest of the graph.)

The test query for 6 asserts that the algorithm will select edge 1, but 6, 7, and 14 are valid (and edge 14 is selected when built against more modern versions of Boost).

Lingua principale
C++
Stelle
1.4k
Fork
397
Merge medio
2g 2h
PR unite (30g)
4

Preparare l'ambiente

Come iniziare

  1. Leggi tutta la issue e poi la guida ai contributi del progetto.
  2. Commenta sulla issue per dire che te ne occupi tu — evita che due persone facciano lo stesso lavoro.
  3. Fai un fork del repository e lavora su un branch.
  4. Apri una pull request che faccia riferimento al numero della issue.

Altre issue di pgRouting/pgrouting

Tutte le issue di pgRouting/pgrouting

Issue simili

Altre issue su C++

Ricevi le nuove issue nella tua casella

Un breve riepilogo di issue GitHub adatte ai principianti.