[Rd] configure --with-pcre1 fails with latest R 4.0 on Ubuntu 14.04
Tomas Kalibera
tom@@@k@||ber@ @end|ng |rom gm@||@com
Sat Mar 21 07:55:58 CET 2020
On 3/18/20 6:11 PM, Hervé Pagès wrote:
> Thanks Tomas. Any chance the old version of the error message could be
> restored? It would definitely be more helpful than the current one.
> It's confusing to get an error and be told to use --with-pcre1 when
> you're already using it.
The message now gives the required version and UTF-8 support
requirement, so one does not have to look that one line up.
Thanks to Brian Ripley,
Tomas
>
>
> H.
>
> On 3/18/20 01:08, Tomas Kalibera wrote:
>> On 3/17/20 8:18 PM, Hervé Pagès wrote:
>>> Using --with-pcre1 to configure the latest R 4.0 (revision 77988) on
>>> an Ubuntu 14.04.5 LTS system gives me the following error:
>>>
>>> ...
>>> checking if lzma version >= 5.0.3... yes
>>> checking for pcre2-config... no
>>> checking for pcre_fullinfo in -lpcre... yes
>>> checking pcre.h usability... yes
>>> checking pcre.h presence... yes
>>> checking for pcre.h... yes
>>> checking pcre/pcre.h usability... no
>>> checking pcre/pcre.h presence... no
>>> checking for pcre/pcre.h... no
>>> checking if PCRE1 version >= 8.32 and has UTF-8 support... no
>>> checking whether PCRE support suffices... configure: error: pcre2
>>> library and headers are required, or use --with-pcre1
>>>
>>> Maybe the real problem is that the PCRE version on this OS is 8.31?
>>
>> Yes, R requires PCRE version at least 8.32 as documented in R-Admin,
>> and this is since September 2019.
>>
>>> The error message is not particularly helpful.
>>
>> An earlier version of the message gave the requirement explicitly,
>> when people would have been more likely to have that old versions of
>> PCRE1.
>> The few who still have it now need to see also the output line above
>> to get the requirement and/or look into the manual.
>>
>> R 4.0 is still keeping support for PCRE1 (>=8.32), but PCRE2 should
>> be used whenever possible.
>>
>> Best,
>> Tomas
>>
>>>
>>> Thanks,
>>> H.
>>>
>>
>
More information about the R-devel
mailing list