Fix/propagate cslc date fmt to ministack#710
Open
scottstanie wants to merge 5 commits intoisce-framework:mainfrom
Open
Fix/propagate cslc date fmt to ministack#710scottstanie wants to merge 5 commits intoisce-framework:mainfrom
scottstanie wants to merge 5 commits intoisce-framework:mainfrom
Conversation
When run_wrapped_phase_sequential is invoked with a non-default cslc_date_fmt (e.g. "%Y%m%d%H%M%S" for non-SSO cadences with same-day repeats), the date format was not forwarded to MiniStackPlanner. As a result, phase-linked SLCs and CRLB rasters were written using the default "%Y%m%d", and create_ifgs in wrapped_phase later raised IndexError trying to extract dates from those filenames with the caller's longer format. Two complementary fixes: 1. sequential.py: pass file_date_fmt=cslc_date_fmt when constructing MiniStackPlanner. 2. stack.py: in MiniStackPlanner.plan(), forward self.file_date_fmt to each MiniStackInfo it creates. The planner already accepted the field but did not propagate it to its children. With both in place, _name_slcs / _name_crlbs produce filenames that preserve the time-of-day component, and create_ifgs can parse them back round-trip. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
After the prior commit, phase-linked SLCs were written with the caller's
cslc_date_fmt (e.g. "%Y%m%d%H%M%S"), but the interferogram VRT names
were still hardcoded to "%Y%m%d" because Network and convert_pl_to_ifg
default to that format. The resulting "%Y%m%d_%Y%m%d.int.vrt" filenames
then failed downstream stitching: stitching_bursts.run calls
merge_by_date with file_date_fmt=cslc_date_fmt, the long-format regex
matches no dates in the date-only filenames, and merge_images either
groups all ifgs together or builds a clip window with negative
dimensions ("Computed -srcwin ... has negative width and/or height").
Forward file_date_fmt into:
- interferogram.Network(...) constructions in create_ifgs (3 sites)
- interferogram.VRTInterferogram(...) construction for the extra-ref
case
- interferogram.convert_pl_to_ifg(...) (2 sites)
convert_pl_to_ifg now accepts a date_format parameter (default unchanged
for backward compatibility) and uses it for both date extraction from
the input SLC name and output filename formatting.
With this and the prior MiniStack propagation in place, IFG VRT
filenames preserve the time-of-day component end-to-end and downstream
merge_by_date round-trips correctly.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Spurt's emcf workflow now accepts a --date-fmt CLI flag (default "%Y%m%d") that controls how acquisition dates are extracted from SLC filenames and written into unwrapped output filenames. Without forwarding the caller's cslc_date_fmt, dolphin's spurt subprocess used the default %Y%m%d slice and same-day acquisitions collapsed to a single date in the output (e.g. 20240709_20240709.unw.tif), losing the time-of-day component already preserved in the upstream ifg names. Thread file_date_fmt: str = "%Y%m%d" through: - unwrap_spurt: append "--date-fmt <fmt>" to the spurt subprocess command, and forward to filled_masked_unw_regions for the ambiguity-interpolation gap-fill step (which already accepted the parameter but was being called with the default). - unwrap.run: forward into unwrap_spurt. - workflows.unwrapping.run: forward into unwrap.run. - workflows.displacement.run: pass cfg.input_options.cslc_date_fmt into unwrapping.run. With this in place, e.g. cslc_date_fmt="%Y%m%d%H%M%S" produces spurt outputs like 20240709040329_20240712025153.unw.tif round-trippable end-to-end. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
timeseries.run already accepted file_date_fmt, but it was not being
forwarded from displacement.run, and several internal call sites
ignored it. As a result, with cslc_date_fmt="%Y%m%d%H%M%S" the
upstream stages produced ifg/unw filenames with time-of-day, but the
timeseries/ outputs collapsed back to %Y%m%d (e.g.
20251002_20251005.tif), losing the disambiguation that motivated the
custom format in the first place.
Forward file_date_fmt at five sites:
- displacement.run: pass cfg.input_options.cslc_date_fmt into
timeseries.run.
- timeseries.run -> invert_unw_network: forward file_date_fmt (the
callee accepted but the caller did not pass it).
- _redo_reference: pass fmt=file_date_fmt into format_dates when
building the re-referenced filenames for extra_reference_date.
- invert_unw_network: pass fmt=file_date_fmt into both format_dates
calls that build out_paths and out_residuals_paths.
- create_nonzero_conncomp_counts: replace hardcoded
d.strftime('%Y%m%d') with d.strftime(file_date_fmt) (the function
already exposed the parameter).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Change Summary
When
run_wrapped_phase_sequentialis invoked with a non-defaultcslc_date_fmt(e.g. "%Y%m%d%H%M%S" with same-day repeats, like Delta-X needs), the date format was not forwarded toMiniStackPlanner. As a result, phase-linked SLCs and CRLB rasters were written using the default "%Y%m%d", and create_ifgs in wrapped_phase later raised IndexError trying to extract dates from those filenames with the caller's longer format.Two fixes:
sequential.py: passfile_date_fmt=cslc_date_fmtwhen constructingMiniStackPlanner.stack.py: inMiniStackPlanner.plan(), forwardself.file_date_fmtto each MiniStackInfo it creates. The planner already accepted the field but did not propagate it to its children.With both in place, _name_slcs / _name_crlbs produce filenames that preserve the time-of-day component, and create_ifgs can parse them back round-trip.
Related issue number
Checklist