IMO we should open an issue to make the regenerate task do some kind of check and fail in a clear way if this happens. It would save people from crazy debugging.
On Sat, Dec 7, 2019 at 11:14 AM Robert Muir <[email protected]> wrote: > Hi Erick. sorry for the slow reply on this one. make sure you have correct > icu4c version at the beginning of your PATH before running ant regenerate. > it should match the icu4j version. it seems to me you have a mismatch. > > On Wed, Dec 4, 2019, 2:32 PM Erick Erickson <[email protected]> > wrote: > >> I have the git pull working for fetching a particular revision of >> nfkc.txt and the like. Now TestICUFoldingFilterFactory fails tests. Here's >> what I could find on that topic: >> >> org.apache.lucene.analysis.icu.ICUFoldingFilter >> public static final Normalizer2 NORMALIZER = Normalizer2.getInstance( >> // TODO: if the wrong version of the ICU jar is used, loading these >> data files may give a strange error. >> // maybe add an explicit check? >> http://icu-project.org/apiref/icu4j/com/ibm/icu/util/VersionInfo.html >> ICUFoldingFilter.class.getResourceAsStream("utr30.nrm"), >> "utr30", Normalizer2.Mode.COMPOSE); >> eventually calls: >> >> com.ibm.icu.impl.Normalizer2Impl >> public Normalizer2Impl load(ByteBuffer bytes) { >> try { >> this.dataVersion = ICUBinary.readHeaderAndDataVersion(bytes, >> 1316121906, IS_ACCEPTABLE); >> which throws >> Caused by: com.ibm.icu.util.ICUUncheckedIOException: java.io.IOException: >> ICU data file error: Header authentication failed, please check if you have >> a valid ICU data file; data format 4e726d32, format version 4.0.0.0 >> >> 0x4e726d32==1316121906, so the data format looks ok to my uninformed eye. >> >> The jar file I have for icu is: icu4j-62.1.jar >> >> I looked at the nfc* files that are now fetched from github and at least >> ./lucene/analysis/icu/src/data/utr30/nfc.txt is identical. >> >> I’ll get back to this later this afternoon, meanwhile any pointers? >> --------------------------------------------------------------------- >> To unsubscribe, e-mail: [email protected] >> For additional commands, e-mail: [email protected] >> >>
