The workaround is simple. The severity is low. Call me shallow, but if I downloaded 0.32 for the first time in order to evaluate monotone, this would probably cause me to put it back on the shelf. Lack of versioned policy, performance, or other lack of features probably would not.
That's not a criticism, suggestion, instruction, or anything of the kind. Just some (annoying) feedback from an end user. ;-) RS On 2/2/07, Christian Ohler <[EMAIL PROTECTED]> wrote:
Thomas Keller, 2007-02-02: > Daniel Atallah schrieb: >> $ mtn list ignored >> mtn: fatal: std::logic_error: paths.cc:308: invariant >> 'I(is_valid_internal(data()))' violated > > Has this already been fixed for 0.33? It's still there. I looked into it briefly about a week ago; fixing it turned out to require more familiarity with the code than I have. Here's what I recall. It's the same issue as the xfail test in ls_unknown_in_subdir. It can be worked around by calling ls unknown/ls missing from the workspace root with a restriction argument instead of calling it from a subdirectory. The bug was introduced by the addition of a call to normalize_external_path in bookkeeping_path::is_bookkeeping_path in 460c36c1128986f68bcf5e9402d2b5abbff19fd5. However, removing that call breaks other tests. When walk_tree_recursive calls bookkeeping_path::is_bookkeeping_path to avoid descending into _MTN, the normalization in bookkeeping_path::is_bookkeeping_path prepends the current working directory to the path; the resulting (bogus) path no longer starts with _MTN and is therefore not detected as a bookkeeping path. After failing to detect that it's in fact looking at _MTN, the code ultimately aborts with the above assertion. I hope this helps. Christian Ohler. _______________________________________________ Monotone-devel mailing list Monotone-devel@nongnu.org http://lists.nongnu.org/mailman/listinfo/monotone-devel
_______________________________________________ Monotone-devel mailing list Monotone-devel@nongnu.org http://lists.nongnu.org/mailman/listinfo/monotone-devel