# HG changeset patch # User Henry S. Thompson # Date 1694517244 -3600 # Node ID e606c609f813d0017db4e2144425bd4c48849ac5 # Parent e2e64c3d763eca110686069865acd3c14a7d475e reinstate better check to start queuing, check for missed match leading to potential runaway diff -r e2e64c3d763e -r e606c609f813 bin/merge_date.py --- a/bin/merge_date.py Mon Sep 11 22:06:45 2023 +0100 +++ b/bin/merge_date.py Tue Sep 12 12:14:04 2023 +0100 @@ -133,41 +133,42 @@ else: # still looking, save if >= date else fall through to write if DEBUG>1: - print("Diso: xkey: %s\n" + print("Diso: match: %s\n" + " xkey: %s\n" " dkey: %s\n" " xdate: %s\n" " ddate: %s\n" " xurl: %s\n" " durl: %s\n" - "xl: %s"%(xkey, dkey, xdate, ddate, xurl, durl, xl), + "xl: %s"%(not noMatch, + xkey, dkey, xdate, ddate, xurl, durl, xl), file=sys.stderr) - if not noMatch: + if (dkey.startswith(xkey1) and + (ddate!=xdate or (xkey2 is not None and dkey!=xkey2))): xq.append(xl) if DEBUG>1: sys.stderr.write('xpush\n') continue # else fall through - else: - # Not messyD - if messyU: - raise ValueError("messyU w/o messyD:" - "xkey: %s\n" - "dkey: %s\n" - "xdate: %s\n" - "ddate: %s\n" - "xurl: %s\n" - "durl: %s\n" - "dfq: %s\n" - "k1, k2: |%s|%s|\n" - "FN: %s XCNT: %s DCNT: %s\n" - "xl: %s"%(xkey, dkey, xdate, ddate, xurl, durl, - (b'\n '.join(dfq)).decode('ascii'), - xkey1, xkey2, FN, XCNT, DCNT, xl)) if (ddate != xdate or not dkey.startswith(xkey1) or (xkey2 is not None and dkey!=xkey2) or durl!=xurl): + if DEBUG and dkey and xkey.decode('ascii')>(dkey.decode('ascii')): + + raise ValueError("Fail: xkey: %s\n" + " dkey: %s\n" + " xdate: %s\n" + " ddate: %s\n" + "dfq: %s\n" + "k1, k2: |%s|%s|\n" + "FN: %s XCNT: %s DCNT: %s\n" + "xl: %s"%(xkey, dkey, xdate, ddate, + (b'\n '.join(dfq)).decode('ascii'), + xkey1, xkey2, FN, XCNT, DCNT, xl)) NF.write(xl) + if DEBUG>1: + sys.stderr.write("out_nl\n") continue # Got it NF.write(xkey) @@ -177,7 +178,7 @@ NF.write(xprops[:-2]) NF.write(b', "lastmod": "%d"}\n'%int(dtime[:-3])) if DEBUG>1: - sys.stderr.write("out: %s"%(' '.join(xp.decode('ascii') + sys.stderr.write("out_t: %s"%(' '.join(xp.decode('ascii') for xp in (xkey, xdate, xurl)))) sys.stderr.write(" %d\n"%int(dtime[:-3])) dl = df.readline()