LAST production releases

This page lists the AstroPack release tags that run in production on the LAST computers. For each tag it gives a short description and the changes since the previous tag (the "delta").

How the tags are used. A release is an annotated tag on main, named lv<major>.<minor>.<patch>. The LAST computers do not follow main: they are fixed at one tag. The tag is set in the LAST_tools file files/github-repos.gledmagicwater (branch:main,tag:<tag>), so last-fetch-from-github keeps them at the tag. A change reaches production only when a new tag is made and that file is updated. See issue #1266. Since lv1.0.8 a tag may instead sit on the maintenance branch lv1.0.x (made from the previous tag), when main holds changes that are not ready for production.

What the numbers mean. The last number (lv1.0.1, lv1.0.2, …) counts fixes: each corrects a specific problem, and products made before and after it are otherwise made the same way. The middle number (lv1.0 → lv1.1) is a step of a higher level: it changes how the products themselves are made — the photometric calibration, the coadd algorithm, the product files — so products of different lv1.x series should not be mixed. lv1.1.0 is such a step. All products made before it were removed from the archive, and all data are reprocessed with lv1.1.0.

lv1.1 series (current)

Tag	Commit	Date	In production	Main changes
lv1.1.0	8e1799119	2026 Oct 19	2026 Oct 19, 10:39–10:41 UT, 19 computers (last08e later)	New series. Photometric calibration no longer drops most calibrators of sharp crops (#1414); visit coadds made differently (#1355, #1166); epoch catalogues as two MAT files per visit (#1417); pipelineII galaxy matching works again (#1422)

lv1.0 series (2026 Oct 11–18)

Tag	Commit	Date	In production	Main changes
lv1.0.10	e6f9650b0	2026 Oct 18	2026 Oct 18, 15:53 UT, 19 computers (last04w off)	bug fix: epoch images have LEVEL = proc, and their image IDs the "proc" level (#1412)
lv1.0.9	6b4ac367e	2026 Oct 18	2026 Oct 18, 14:21 UT, 19 computers (last04w off)	bug fix: the background FITS writer keeps the image until the file is written; write failures are logged (#1391)
lv1.0.8	35c3e6ee5	2026 Oct 18	2026 Oct 18, 13:32 UT, 19 computers (last04w off)	pipelineII reuses the demon's asteroid orbits (less memory); FITS files closed after an error (#1392)
lv1.0.7	ffcba2a5b	2026 Oct 17	2026 Oct 17, 19:37 UT, 19 computers (last04w off)	the raw-image noise check measures 24 boxes across the image, not only the centre (LAST_issues #242)
lv1.0.6	f9c0e292a	2026 Oct 17	2026 Oct 17, 11:39 UT, 19 computers (last04w off)	bug fix: forced-photometry flux errors include the source photon noise (#1408)
lv1.0.5	477959a8e	2026 Oct 16	2026 Oct 16, 22:45 UT, 19 computers (last04w off)	hyperbolic orbits no longer fail the known-asteroid matching (#1410)
lv1.0.4	545f8492a	2026 Oct 15	2026 Oct 15, 19:40 UT, 19 computers (last04w off)	pipelineI only for visits older than 2 days (reprocessing of archived data)
lv1.0.3	71cbf5380	2026 Oct 14	2026 Oct 14, 16:02 UT, all 20 computers	much less memory on very dense fields: Gaia stars for the transient match read per sub-image (#1387)
lv1.0.2	26ba4efe0	2026 Oct 14	2026 Oct 14, 10:26 UT, all 20 computers	correct barycentric JD and velocity in the transients database (#1395)
lv1.0.1	4ec180318	2026 Oct 13	2026 Oct 13, 19:35 UT, all 20 computers	transients written to the database again (#1396); bright-sky images no longer rejected by the histogram check (#1179)
lv1.0.0	456c80f68	2026 Oct 11	2026 Oct 11, 18:10 UT (18 computers), last03w/last04w on Oct 13	first release

lv1.1.0

A new series, one level above the ten lv1.0.x fixes. lv1.1.0 changes how the main products are made: the photometric calibration, the visit coadds and the files of the epoch catalogues. Products made with lv1.0.x and with lv1.1.0 are therefore not directly comparable. All products made before lv1.1.0 were deleted from euclid /bigdata3 and from the LAST computers on 2026 Oct 18–19, and the archived data are reprocessed with lv1.1.0.

Commit 8e1799119 on main, 2026 Oct 19. It contains everything on main since lv1.0.7, including the fixes of lv1.0.8–lv1.0.10 (made on the branch lv1.0.x). Release note: #1266.

What changes in the products

• Photometric calibration (PhotCalibTrans, fitPhotCalibTrans): calibrator stars are no longer rejected for the CR_DeltaHT flag (#1414). Since #1331 that flag lands on nearly every bright star of a sharply focused crop, so such crops lost most of their calibrators and fell back to the reduced (zero-point only) calibration. In the test, reduced-mode crops went from 40 to 1 on 8 visits, and the affected crops have several times more calibrators (e.g. 3 → 47) and smaller residuals; the median zero point is unchanged (within 0.003 mag). (Dana)
• Visit coadds (pipeline.generic.procCoadd, imProc.stack.coaddHeader): the epochs are brought to the same sky level before the outlier rejection, and the new keyword MAXBCKD gives the largest background difference of the coadded epochs (#1355); the coadd variance is scaled by the number of epochs per pixel (#1166). In the test the coadds have 1–6% fewer sources and a limiting magnitude about 0.01 mag shallower; magnitudes agree within about 0.001 mag, except where the sky changed between the epochs (one visit: 0.013 mag).
• Epoch catalogues (pipelineI, imProc.io.saveProductMat): the 480 proc_Cat FITS files of a visit (20 epochs × 24 crops) are replaced by two MAT files, proc_Cat.mat and proc_Head.mat (#1417). The readers (queryDB.loadProducts/loadTable, the DB ingestion) read both forms. The visit gets faster to save and smaller. Approved by Eran and Chen.
• Known asteroids are matched after the photometric calibration (#1423).

Fixes

• pipelineII failed in the galaxy matching on every visit since lv1.0.5: catsHTM.cone_search selected more columns than the PGC catalogue files have (#1422, a regression of #1403).
• All fixes of lv1.0.6–lv1.0.10 are included (#1408, LAST_issues #242, #1392, #1391, #1412); see their sections below.
• Smaller: streak files store ORIGSEC/LIGHTSEC/CROPID (#1415); sliceGaussianProfile slice range clamped (#1389); maskSaturated (#1426), AstroImage.maskSet (#1427), astrometryRefine with catalogue input (#1055); pipelineII requests only the Gaia columns it uses (#1403).

Other changes (not used by the production pipeline)

• DB ingestion and benchmark tools (pipeline.last.insertDB), Installer.installCatsHTM (#1425), nightReport, ds9_crops (#958), stellar-type helpers (astro.stars), ultrasat.simulateGaiaField.

How it was tested. 8 visits from 7 cameras, full pipeline (pipelineI and pipelineII), against lv1.0.10 and against the intermediate commits b762de0cf and b6dc3c4c1: a well-focused October night, a sharp April night, two N2 fields, the dense field 524, a soft-seeing camera, a moonlit visit, the field-841 visit of #1410, and a live visit of 2026 Oct 18/19. pipelineI succeeded on all; pipelineII on 7 of 8 (the 8th is the known #1405, as before). pipelineI is 3–7% slower; peak memory unchanged (59–65 GB). The final commit gives products identical to the tested candidate.

Deployment. 2026 Oct 19, with the pipelines stopped: all 20 computers moved to the tag; pipelines restarted at 10:39 UT (unit 1) and 10:41 UT (unit 2: simultaneous worker-pool starts can fail), last08e later; the first visits processed were those of the night of Oct 18/19. LAST_tools names lv1.1.0 since c9bc812 (last-tool 1.0.9788.19555).

lv1.0.10

Commit e6f9650b0 on the branch lv1.0.x, 2026 Oct 18: lv1.0.9 plus 1 commit (on main as 1c27e5850). Nothing else from main is included. Release note: #1266.

Products change only in three header values. In the epoch catalogs (proc_Cat, image header) LEVEL is proc instead of raw, and the level part of ID_PROC is "proc" (3) instead of "raw" (2); the coadd headers carry the corrected ID_PROC, ID_PROCF and ID_PROCL. ID_COADD, all other header values, all data and all file names are unchanged. Products made before lv1.0.10 keep LEVEL = raw; they are not corrected, because they will be reprocessed.

Changes that affect production processing

• pipeline.last.pipes.pipelineI: sets LEVEL = proc in the epoch images before imProc.db.generateImageID makes ID_PROC (issue #1412). The v0 pipeline already did this in pipeline.generic.singleRaw2proc.

Tested on R2020b on a full visit with lv1.0.9 and lv1.0.10: 689 files, data byte-identical, only the values above differ; checked again on the first production visit on last01e.

Deployment. last01e was moved first (15:43 UT on 2026 Oct 18), the other 18 reachable computers at 15:53 UT, with the pipelines running; each restarted at the end of its current visit (only pipelineI changed, so no visit failed). last04w was off and is still on lv1.0.3; it needs lv1.0.10 with its pipelines stopped first. LAST_tools names lv1.0.10 since afcd171. The later last-tool 1.0.9787.29268 (LAST_tools 1d6f694) also makes the pipeline service wait for the data disks after a reboot: on 2026 Oct 18 a power cut rebooted six computers, and on last01e camera 01 MATLAB started before the disk with the star catalogs was mounted, so 8 visits failed until the next restart (their 160 raw images were moved back to new/).

lv1.0.9

Commit 6b4ac367e on the branch lv1.0.x, 2026 Oct 18: lv1.0.8 plus 1 commit (on main as 5ff75b590). Nothing else from main is included. Release note: #1266.

Products are unchanged (normal output files are byte-identical to lv1.0.8).

Changes that affect production processing

• io.fits.writeThreadMexFITS, io.fits.DataKeeper, io.fits.mex.mex_fits_write_image_thread: the image handed to the background writer is kept until the writer reports that the file is written. Before, it was freed after 30 s even if writing was still going on, which could give a corrupt file or a crash on a slow disk (issue #1391). A write still running after 60 s is reported once in the service log. The writer now also reports failures (cannot open, short write such as a full disk, close error) with the file name; before, a write to a full disk counted as success. Messages go to stderr, because the AstroPack startup turns MATLAB warnings off. Only the Linux mex was rebuilt.

Tested on R2020b against the old writer: single/uint16/int32 files byte-identical; a write blocked for 70 s gave a completely wrong file with the old code (image freed at 30 s) and a correct file with the new one; failed open and /dev/full are reported.

Deployment. last01e was moved first (14:01 UT on 2026 Oct 18), the other 18 reachable computers at 14:15 UT, with the pipelines running; each restarted at the end of its current visit (all 38 by 14:21 UT). Because the fix changes a function and its caller together, a pipeline that was in the middle of a visit at the switch failed to save it ("Too many input arguments"): 3 visits (01.01.01, 01.03.02, 01.05.03), whose 60 raw images were moved back to new/. A release that changes a function and its caller together must be deployed with the pipelines stopped. last04w was off and is still on lv1.0.3. LAST_tools names lv1.0.9 since f3e8235 (last-tool 1.0.9787.25866).

lv1.0.8

Commit 35c3e6ee5 on the branch lv1.0.x, 2026 Oct 18: lv1.0.7 plus 2 commits (the same changes are on main as f935be42b and 8b1af1566). Nothing else from main is included; in particular the photometric-calibration change of #1414 is not in this release. Release note: #1266.

Products are unchanged.

Changes that affect production processing

• pipeline.last.pipes.pipelineII: new argument OrbEl (default [] = load the 'merge' orbital elements as before); PipelineDemon now passes the set it loaded at start, so pipelineII no longer reloads ~1.5 million orbits every visit (about 0.4 GB and a few seconds per visit). Comets are still loaded in pipelineII (issue #1392). (63bbdabcd)
• FITS.read1, FITS.writeTable1: the file is closed when an error happens between opening and closing it (before, each such error left one file open); writeTable1 also returns to the original folder. (35c3e6ee5)

Tested on R2020b with one local visit (field 1488) on lv1.0.7 and on lv1.0.8: all catalogs (epoch, coadd and subtraction, including the asteroid matches) byte-identical; other files differ only in write dates. Forced-error tests: no file left open (lv1.0.7: one per error). FITS.unitTest passes.

Deployment. last01e was moved first (13:16 UT on 2026 Oct 18), the other 18 reachable computers at 13:24 UT. Each pipeline restarted at the end of its current visit (all 38 between 13:17 and 13:32 UT). Daytime visits were reprocessing of old nights, so pipelineII was skipped (older than 2 days); of the first 80 visits 79 were processed and one failed because the parallel pool did not start (last02w, re-queued). last04w was off and is still on lv1.0.3. LAST_tools names lv1.0.8 since 63bf5d5 (last-tool 1.0.9787.24143).

lv1.0.7

Commit ffcba2a5b on main, 2026 Oct 17, 3 commits after lv1.0.6. Release note: #1266.

Changes which raw images pipelineI accepts. Noisy raw images that passed the pixel-noise check before are now rejected. Products of accepted images are unchanged.

Changes that affect production processing

• imProc.quality.noiseExcess (used by pipeline.generic.prePrep): the ratio of measured to expected pixel-to-pixel noise is now the worst of a 4x6 grid of 500x500 boxes over the illuminated region (new arguments Nbox, default [4 6], and CCDSEC, default [1 6388 25 9600]), instead of one central box; MaxRatio unchanged (4). On LAST.01.06.03 in 2026 Jun-Sep the noise covered the whole sensor but was weakest at the centre, so about 1 in 5 noisy frames passed and gave visits with 35,000-45,000 false sources per sub-image (LAST_issues #242). The NoiseRatio value in the image table is now the worst-box value. (ffcba2a5b)

Changes not used by the production pipeline

• astro.stars.tlogg2picklesClass: log(g) dwarf boundary as an argument, default 4.0 (was >4.5). (eafad117d)
• New astro.stars.bprp2teff: Teff and log(g) from Gaia BP-RP, G and parallax (ULTRASAT). (f16b251da)

Tested on 4,319 set-aside 01.06.03 frames: 4,206 rejected (the old check: 3,386). Normal frames on 38 cameras (3,175, dense fields included) give at most 1.23. 0.25 s per frame. prePrep on whole visits: a normal 01.06.03 visit keeps all 20 frames; a noisy 01.05.02 visit is now rejected (worst box 5.1, centre 3.7).

Deployment. last01e was moved first (19:10 UT on 2026 Oct 17; pipelines restarted 19:10 and 19:13 UT); the other 18 reachable computers at 19:28 UT. Each pipeline restarted at the end of its current visit (all 38 between 19:10 and 19:37 UT); the first 29 visits had no failures. last04w was off and is still on lv1.0.3. LAST_tools names lv1.0.7 since 9284d7a (last-tool 1.0.9786.35087).

lv1.0.6

Commit f9c0e292a on main, 2026 Oct 17, 1 commit after lv1.0.5. Release note: #1266.

Bug fix that changes pipelineI products. The forced-photometry rows of the epoch catalogues had magnitude errors and S/N (MAGERR_PSF, SN; the catalogues do not store FLUXERR_PSF) without the source's own photon noise: too small for bright stars, by 1.5x at S/N 30-50, 2.4x at S/N 50-100 and 13x at S/N 200-400. All products made before lv1.0.6 (including the April 2026 reprocessing up to 2026 Oct 17) carry these errors in their forced-photometry rows. The regular PSF photometry of detected sources is not affected.

Changes that affect production processing

• imProc.sources.forcedPhotNew (used by pipelineI) and imProc.sources.forcedPhot: default UseSourceNoise is 'last' (was 'off'); new Gain/KeyGain arguments, the gain is read from the image header (1 if missing) and passed to imUtil.sources.psfPhotCube. Forced-photometry fluxes change by about 0.1-0.3% (#1408). (f9c0e292a)
• pipeline.last.phot.forcedPhot: default UseSourceNoise is 'last'. (f9c0e292a)

Tested on a 20-epoch visit (LAST.01.08.03, 2023 Jun 16, about 10,500 stars): the reported errors match the epoch-to-epoch scatter to within 5% up to S/N 50; an image copy in ADU with GAIN=2 gives the same fluxes and errors.

Deployment. last01e was moved first (10:50 UT on 2026 Oct 17); the other 18 reachable computers at 11:32 UT. Each pipeline restarted at the end of its current visit (all 38 between 10:53 and 11:39 UT); the first 33 visits had no failures. last04w was off and is still on lv1.0.3. LAST_tools names lv1.0.6 since b71dc1d (last-tool 1.0.9786.20772).

lv1.0.5

Commit 477959a8e on main, 2026 Oct 16, 5 commits after lv1.0.4. Release note: #1266.

Changes that affect production processing

• celestial.Kepler.xyz2elements and celestial.OrbitalEl.nu2posOrbitalFrame: correct results for hyperbolic and parabolic orbits (they gave complex numbers and a reversed velocity). Elliptic orbits are unchanged. Before this, the one hyperbolic orbit in the catalogue (2024 UQ) made every visit near its predicted position fail in pipelineI (#1410). (477959a8e)
• pipeline.last.pipes.PipelineDemon: new setting MaxOrbEccen (default 1); orbits with eccentricity of 1 or more are left out of the known-asteroid matching. (477959a8e)
• pipeline.last.pipes.pipelineI: the known-asteroid matching runs per sub image; if it fails, that sub image is saved without known-asteroid matches and a warning is logged, instead of failing the visit. (477959a8e)
• catsHTM.cone_search: trims each catalogue cell to the cone before joining them (less memory and time; same results); new options Method and Columns (#1403). (Dana, c44e499a2)

Tested on 01.09.03 (night 2026-04-05): the field-841 visit that failed in production succeeds; a normal visit with lv1.0.4 and with lv1.0.5 gives identical products (596 files) and the same list of matched known asteroids.

Other changes (not used by the production pipeline)

• pipeline.last.reference.buildRefImages: no references from too few epochs (MinEpoch, #1404). (Dana, b719d0ded)
• PhotCalibTrans: CCDSEC reports the header's section, never a fallback (#1393). (Dana, f1b6d4938)
• nightReport: per-crop calibrator statistics. (Dana, 330e757eb)

Deployment. The 19 reachable computers were moved to the tag at 22:43 UT on 2026 Oct 16; each pipeline restarted at the end of its current visit. last04w was off and is still on lv1.0.3. LAST_tools names lv1.0.5 since e9581a3 (last-tool 1.0.9785.40920).

lv1.0.4

Commit 545f8492a on main, 2026 Oct 15, 7 commits after lv1.0.3 (plus 1 merge). Release note: #1266.

Changes that affect production processing

• pipeline.last.pipes.PipelineDemon has a new setting PipelineIIMaxAgeDays (default Inf, no limit). A visit whose coadd MIDJD is older than this gets pipelineI only; pipelineII is skipped, the skip is logged ("Pipeline II skipped: visit is N days old"), and the products are still marked ready for transfer. startBatchLASTPipeline sets 2 days, so live visits keep pipelineII. Made for the reprocessing of archived raw images. (545f8492a)
• imProc.stack.coadd_WRobust: the coadd header keyword COADDOP is wrobust (was the typo wrbosut). (1eabcabce)
• tools.math.stat.rstd: a faster method for NaN-free vectors; results are identical (#1377). (Ruslan, 399e48b20)

Tested on 3 visits (01.01.02, 01.04.01, 01.07.01) with lv1.0.3 and the new code: the only difference is COADDOP in the coadd image and catalog files. A 36-day-old visit run with the 2-day limit skipped pipelineII, gave the same pipelineI products as the full run, and took 312 s instead of 370 s.

Other changes (not used by the production pipeline)

• PhotCalibTrans / fitPhotCalibTrans: a selectable filter-curve store (FilterSource), and VO.ZTF reads a ZTF per-epoch catalog in the LAST format, for calibrating other missions (#1393). The default is unchanged. (Dana, 824a12347, 78e4f6726)
• nightReport: a short summary of the night's problems, units on all plots, logarithmic histograms and a field check. (Dana, 41e1502c6)
• imUtil.properCoadd.unitTest follows the properCoaddLinearR -> properCoaddLinear rename. (Dana, 883a98bc1)

Deployment. The 19 reachable computers were moved to the tag at 19:40 UT on 2026 Oct 15 and all 38 pipelines were restarted while idle. last04w was off and is still on lv1.0.3 until it is back. LAST_tools names lv1.0.4 since 39d1af6 (last-tool 1.0.9784.35482).

lv1.0.3

Commit 71cbf5380 on main, 2026 Oct 14, 4 commits after lv1.0.2 (plus 2 merges). Release note: #1266.

Changes that affect production processing

• pipeline.last.pipes.pipelineII reads the Gaia stars for matching transient candidates for each sub-image (a cone of the sub-image half-diagonal + 300") instead of one catalog for the whole visit. Near the Galactic Centre the visit-wide query held 8.4 million stars and briefly needed about 20 GB more memory. Tested on 7 visits with lv1.0.2 and the new code: all products identical; peak memory 87.7 -> 63.5 GB (field 524) and 80.2 -> 63.5 GB (field 608); a normal field is not slower (#1387). (Ruslan, 0d7a36375)

Other changes (not used by the production pipeline)

• imProc.stack.coadd_Proper: missing pixels are no longer counted as real measurements (#1399). (Dana, 3d85f884a); imUtil.properCoadd.properCoaddLinear updated. (Eran, 5398854eb) Production coadds use coadd_WRobust.
• pipeline.last.reference.buildRefImages and imProc.stack.stitchCrops: reference building recovers the node number of old coadds and finds coadds named with a different time. (d5268dd84)

Deployment. The computers were moved to the tag at 16:01 UT on 2026 Oct 14 and each pipeline was restarted when idle (39 of 40 by 16:04 UT; 01.06.03 at the end of the visit it was processing). LAST_tools names lv1.0.3 since 607e04a (last-tool 1.0.9783.28891). The newer commit 399e48b20 (#1377) was not tested and is not in this release.

lv1.0.2

Commit 26ba4efe0 on main, 2026 Oct 14, 3 commits after lv1.0.1.

Changes that affect production processing

• celestial.time.barycentricJD gave wrong results whenever it was called with more than one time or position: the light-travel-time correction and the barycentric velocity were divided by about the square root of the number of rows. imProc.db.insertCatalog calls it for a whole catalog at once, so the BJD and BARYVEL columns written to the database were wrong (in last.diff_src about 5 times too small). Single-value calls were always correct. Rows already in the database are not corrected by this release (#1395). (26ba4efe0)

Other changes (not used by the production pipeline)

• pipeline.generic.procCoadd: a proper coadd is now always treated as background-subtracted by the source extractor (#1394). Production coadds use coadd_WRobust. (Dana, 3a79cd2bc)
• imProc.calib.fitPhotCalibTrans: bands of other missions can be calibrated through a Mission setting (#1393). The default mission is LAST, and the LAST results do not change. (Dana, 394bcdf90)

Deployment. Deployed on 2026 Oct 14 with the pipelines stopped and restarted: last01e at 10:21 UT, the other 19 computers at 10:26 UT. LAST_tools names lv1.0.2 since 6338aa1 (last-tool 1.0.9783.18831).

lv1.0.1

Commit 4ec180318 on main, 2026 Oct 13, 24 commits after lv1.0.0 (plus 5 merges). Release note: #1266.

Changes that affect production processing

• Transient candidates are written to the database again (#1396). Nothing had been written since 2026 Sep 18. Three code problems were fixed:
  • transient writing was not turned on in production;
  • the native ClickHouse connector failed on every insert (empty schema, numbers in the wrong column type);
  • NaN values made the insert fail. Re-reductions of visits older than 2 days do not write. (bbed03abf, 0670d4633)
• The histogram check (imUtil.image.histAnomaly) is skipped when the sky is above 3000 ADU. On a bright (moonlit) sky it gave false alarms and rejected good images (#1179). (4ec180318)
• imProc.sub.smearTemplate (used by pipelineII) uses the derived smear in preference to the measured one, and its S/N column is fixed. (Ruslan, 6da0e0ad2)

Other changes (not used by the production pipeline)

• imProc.stack.coadd_Proper: the background is measured on the normalized coadd scale (#1329), and the PSF star count is set so the seeing is measured (#1390). Production coadds use coadd_WRobust. (Dana)
• New ts.fitps functions (power-law and damped-random-walk power-spectrum fits) and ts.rand.randomPowerLawLightCurve. (Eran)
• Database insert speed tests: imProc.db.insertCatalog gets a SkipDbInsert option (default false), plus a separate insertCatalog_perf, performance-logging functions, scripts and notes under pipeline.last.insertDB. (Chen)

Deployment. 0670d4633 (the #1396 fix) was deployed on 2026 Oct 13, 18:55 UT. 4ec180318 followed at 19:29–19:35 UT, and the tag was then made on that same commit. LAST_tools names lv1.0.1 since 6f46e44 (last-tool 1.0.9782.35472).

lv1.0.0

Commit 456c80f68 on main, 2026 Oct 11. This is the first release (#1266): the code that dev1 had when it was merged into main on 2026 Sep 26, with later fixes up to Oct 11. It runs the v1 pipeline (PipelineDemon) on all cameras. Its last changes before the tag:

• the memory and file-handle leaks in the FITS writer, the FITS readers and sliceGaussianProfile are closed (#1388);
• pipelineI frees the full-frame images after the cut into sub-images, about 9 GB less memory (#1367);
• pipelineI runs maskHoles, fwhm and fitPhotCalibTrans on the parallel workers, about 25 s faster per visit (#1256).

There is no earlier tag to compare it with. For the changes made before Oct 11 see the New functions and changes log.