swmmio

swmmio copied to clipboard

Loading outfalls seems to crash depending on the order of the free and timeseries staged outfalls. Perhaps a bug?

Loading these outfalls works fine:`

[OUTFALLS]
;;Name           Elevation  Type       Stage Data       Gated    Route To        
;;-------------- ---------- ---------- ---------------- -------- ----------------
StagedOutfall    0          TIMESERIES OutfallStage     NO                       
FreeOutfall      0          FREE                        NO                       

but these outfalls raise an IndexError:

[OUTFALLS]
;;Name           Elevation  Type       Stage Data       Gated    Route To        
;;-------------- ---------- ---------- ---------------- -------- ----------------
FreeOutfall      0          FREE                        NO                       
StagedOutfall    0          TIMESERIES OutfallStage     NO                       

The error is:

IndexError: failed to parse [OUTFALLS] with cols: ['Name', 'InvertElev', 'OutfallType', 'StageOrTimeseries']. head:
[OUTFALLS]


FreeOutfall      0          FREE                        NO
StagedOutfall    0          TIMESERIES OutfallStage     NO

This unittest fails for the inp file where the free outlet is first:

import os
import unittest
import swmmio

class SwmmioTest(unittest.TestCase):

    def test_swmmio_outfall(self):
        
        swmm_file = os.path.join('path to test files', 'outfalls_issue.inp')
        model = swmmio.Model(swmm_file)

        try:        
            outfalls = model.inp.outfalls
        except:
            self.assertTrue(False, 'Could not read outfalls from inp file')

        swmm_file_free_first = os.path.join('path to test files', 'outfalls_issue_free_first.inp')
        model_free_first = swmmio.Model(swmm_file_free_first)

        try:        
            outfalls_free_first = model_free_first.inp.outfalls
        except:
            self.assertTrue(False, 'Could not read outfalls from inp file when free outfall was first')

The two inp file both run in SWMM5.1 (.txt needs to be removed): outfalls_issue.inp.txt outfalls_issue_free_first.inp.txt

I am on Windows 11, python 3.9.7, pandas 1.5.1 and using swmmio version 0.6.2.

@MortenGrum thank you for identifying this issue with so much detail! I'm going to investigate this over the next day or two. Hopefully we can leverage your unit test to fix this.

Hey @aerispaha, this is the same issue I mentioned to you some time ago. Are you planning on merging this fix? Thank you @BuczynskiRafal. This addresses a bunch of issues that happens when data blocks have rows with more elements that the first row.

Hi @lmontest! I would like to merge this fix. I'm having trouble constructing the PR though. This commit on @BuczynskiRafal 's fork seems to be in a unusual state, with GitHub showing this warning:

[!WARNING] This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.

@BuczynskiRafal can you advise? If you can create a PR with these changes, I'll review it asap.

I think this warning came up because I closed PR. I closed PR because @aerispaha wrote that he was already working on a more general solution to this issue. I can restore the PR but I'm afraid it didn't solve the problem completely because the whole test suite didn't pass.