#401 ongoing: add icudt 69.1

.Rbuildignore

@@ -1,15 +1,15 @@
 ^NEWS_TODO$
+^src/icu[0-9]+/data
 ^.*\.Rproj$
 ^\.Rproj\.user$
 ^.*\.kdev4$
 ^\.kdev4
 ^devel
+kate-swp$
 ^src/boost
 ^src-i386
 ^src-x64
 ^src/.*\.o$
-^src/icu55/data
-^src/icu61/data
 ^src/.*\.a$
 ^src/.*\.so$
 ^src/.*\.dll$

.github/workflows/r-icu-bundle.yml

@@ -26,3 +26,4 @@ jobs:
         sudo make r-icu-bundle
         sudo make tinytest
         sudo make check
+        sudo make check-cran

.gitignore

@@ -17,6 +17,7 @@ config.status
 .Rhistory
 .RData
 stringi.Rcheck
+*.kate-swp
 
 # compiled/precompiled files
 *.o
@@ -44,11 +45,4 @@ src-x64/
 cache-knitr/
 cache/
 
-# benchmark results and data
-devel/benchmarks/results*
-devel/benchmarks/figure
-devel/benchmarks/report-*.md
-devel/benchmarks/report-*.html
-devel/benchmarks/test1.csv.gz
-
 .vscode

Makefile

@@ -6,9 +6,6 @@
 
 all: r
 
-#CPPFLAGS="-fopenmp -march=native -mtune=native"
-#LDFLAGS="-fopenmp"
-
 autoconf:
 	autoconf
 	Rscript -e 'roxygen2::roxygenise(roclets=c("rd", "collate", "namespace", "vignette"), load_code=roxygen2::load_installed)'
@@ -35,6 +32,10 @@ build:
 check: build
 	cd .. && R CMD check `ls -t stringi*.tar.gz | head -1` --no-manual #--as-cran
 
+check-cran: build
+	cd .. && STRINGI_DISABLE_PKG_CONFIG=1 R CMD check `ls -t stringi*.tar.gz | head -1` --as-cran
+
+
 weave:
 	cd devel/sphinx/weave && make && cd ../../../
 
@@ -56,7 +57,9 @@ sphinx: r weave rd2rst news
 	touch docs/.nojekyll
 
 clean:
-	rm -f src/*.o src/*.so src/Makevars src/uconfig_local.h \
+	find src -name '*.o' -exec rm {} \;
+	find src -name '*.so' -exec rm {} \;
+	rm -f src/Makevars src/uconfig_local.h \
 		src/install.libs.R config.log config.status src/symbols.rds
 
 purge: clean

NEWS

@@ -8,32 +8,35 @@
     #405 (stri_sub match.length attrib)
 
 
-
 ## 1.6.1 (2021-XX-YY) **devel**
 
+* [GENERAL] #401: stringi is now bundled with ICU4C 69.1 (upgraded from 61.1),
+  which is used on most Windows and OS X builds as well as on *nix systems
+  not equipped with system ICU. However, if the C++11 support is disabled,
+  stringi will be built against the battle-tested ICU4C 55.1.
+  The update to ICU brings Unicode 13.0 and CLDR 39 support.
+
+* [GENERAL] stringi now requires R >= 3.1 (`CXX_STD=CXX11`).
+
 * [DOCUMENTATION] A draft version of a paper on `stringi` is now available at
   https://stringi.gagolewski.com/_static/vignette/stringi.pdf
 
-* ...todo... #401 (update ICU4C to 69.1),
-    The ICU4C bundle has been updated from version 61.1 to 69.1
-    which features Unicode 13.0 and CLDR 39.
-
 * [NEW FEATURE] #408: `stri_trans_casefold()` performs case folding;
-    this is different from case mapping, which is locale-dependent.
-    Folding makes two pieces of text that differ only in case identical.
-    This can come in handy when comparing strings.
+  this is different from case mapping, which is locale-dependent.
+  Folding makes two pieces of text that differ only in case identical.
+  This can come in handy when comparing strings.
 
 * [NEW FEATURE] #421: `stri_rank()` ranks strings in a character vector
-    (e.g., for ordering data frames with regards to multiple criteria,
-    the ranks can be passed to `order()`, see #219).
+  (e.g., for ordering data frames with regards to multiple criteria,
+  the ranks can be passed to `order()`, see #219).
 
 * [BUGFIX] `stri_sort_key()` now outputs `bytes`-encoded strings.
 
 * [BUGFIX] #415: `locale=''` was not equivalent to `locale=NULL`
-    in `stri_opts_collator()`.
+  in `stri_opts_collator()`.
 
 * [INTERNAL] #414: Use `LEVELS(x)` macro instead of accessing `(x)->sxpinfo.gp`
-    directly (@lukaszdaniel).
+  directly (@lukaszdaniel).
 
 
 ## 1.5.3 (2020-09-04) **CRAN**

R/install.R

@@ -31,55 +31,40 @@
 ## EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 
 
-# @title
-# Installation-Related Utilities [DEPRECATED]
-#
-# @description
-# These functions are responsible for checking and guaranteeing
-# that the ICU data library (icudt) is available and that \pkg{stringi}
-# is ready to use.
-#
-# These functions are deprecated and will no longer be available
-# in future \pkg{stringi} releases.
-#
-# @details
-# ICU makes use of a wide variety of data tables to provide many
-# of its services. Examples include converter mapping tables,
-# collation rules, transliteration rules, break iterator rules
-# and dictionaries, and other locale data.
-#
-# Without the ICU data library (icudt) many \pkg{stringi} features
-# will not be available. The size of icudt is approx. 10-30 MB.
-#
-# \code{stri_install_check()} tests whether some ICU services
-# are available. If this is not the case, it is most likely due to
-# unavailable ICU data library.
-#
-# \code{stri_install_icudt()} downloads and installs the ICU data library
-# specific to your platform (little/big-endian). The downloaded
-# file will be decompressed into the directory where the package has been
-# installed, see \code{\link{find.package}}, so make sure
-# you have sufficient write permissions.
-#
-# @param silent suppress diagnostic messages
-# @param check enable \code{stri_install_check()} tests
-# @param outpath path to install icudt to. If \code{NULL}, then
-# \code{file.path(path.package('stringi'), 'libs')} will be used.
-# @param inpath path to search icudt archive in.
-# If \code{NULL}, then only stringi mirror servers will be used.
-# Mainly of interest to system admins and software developers.
-#
-# @return These functions return a logical value, invisibly.
-# \code{TRUE} denotes that the requested operation has been completed
-# successfully.
-#
-# @references
-# \emph{ICU Data} -- ICU User Guide,
-# \url{http://userguide.icu-project.org/icudata}
-#
-# @examples
-# stri_install_check()
-#
+# internal functions used whilst installing stringi
+
+
+
+icudt_fname <- c(
+    little55 = "icudt55l.zip",
+    big55 = "icudt55b.zip",
+    little61 = "icudt61l.zip",
+    big61 = "icudt61b.zip",
+    little69 = "icu4c-69_1-data-bin-l.zip",
+    big69 = "icu4c-69_1-data-bin-b.zip"
+)
+
+icudt_md5ex <- c(
+    little55 = "ff345529f230cc39bb8d450af0607708",
+    big55 = "1194f0dd879d3c1c1f189cde5fd90efe",
+    little61 = "6d14e059b26606f08bad3b41eb3b5c93",
+    big61 = "45719f3208b2d67132efa620cecccb56",
+    little69 = "58ecd3e72e9d96ea2876dd89627afeb8",
+    big69 = "e86eba75d1f39be63713569dc0dc9524"
+)
+
+
+# icudt_mirrors <- c(
+# #     "https://github.com/unicode-org/icu/releases/download/release-69-1/",
+#     "https://raw.githubusercontent.com/gagolews/stringi/master/src/icu69/data/",
+#     "https://raw.githubusercontent.com/gagolews/stringi/master/src/icu61/data/",
+#     "https://raw.githubusercontent.com/gagolews/stringi/master/src/icu55/data/",
+#     "http://raw.githubusercontent.com/gagolews/stringi/master/src/icu69/data/",
+#     "http://raw.githubusercontent.com/gagolews/stringi/master/src/icu61/data/",
+#     "http://raw.githubusercontent.com/gagolews/stringi/master/src/icu55/data/"
+# )
+
+
 # @rdname stri_install
 stri_install_check <- function(silent = FALSE)
 {
@@ -115,38 +100,19 @@ stri_install_check <- function(silent = FALSE)
 }
 
 
-
-
-icudt_fname <- c(
-    little55 = "icudt55l.zip",
-    big55 = "icudt55b.zip",
-    little61 = "icudt61l.zip",
-    big61 = "icudt61b.zip")
-
-icudt_md5ex <- c(
-    little55 = "ff345529f230cc39bb8d450af0607708",
-    big55 = "1194f0dd879d3c1c1f189cde5fd90efe",
-    little61 = "6d14e059b26606f08bad3b41eb3b5c93",
-    big61 = "45719f3208b2d67132efa620cecccb56")
-
-icudt_mirrors <- c("https://raw.githubusercontent.com/gagolews/stringi/master/src/icu61/data/",
-    "https://raw.githubusercontent.com/gagolews/stringi/master/src/icu55/data/",
-    "http://raw.githubusercontent.com/gagolews/stringi/master/src/icu61/data/",
-    "http://raw.githubusercontent.com/gagolews/stringi/master/src/icu55/data/",
-    "http://www.ibspan.waw.pl/~gagolews/stringi/",
-    "http://www.gagolewski.com/software/stringi/")
-
-
-
 # @rdname stri_install
 stri_download_icudt <- function(inpath, icu_bundle_version)
 {
-
     fname <- icudt_fname[paste0(.Platform$endian, icu_bundle_version)]
 
     md5ex <- icudt_md5ex[paste0(.Platform$endian, icu_bundle_version)]
 
-    mirrors <- icudt_mirrors
+#     mirrors <- icudt_mirrors
+    mirrors <- sprintf(
+        "%s://raw.githubusercontent.com/gagolews/stringi/master/src/icu%d/data/",
+        c("https", "http"),
+        icu_bundle_version
+    )
 
     icudtzipfname <- file.path(inpath, fname)  #tempfile(fileext='.zip')
 
@@ -173,7 +139,7 @@ stri_download_icudt <- function(inpath, icu_bundle_version)
         tryCatch({
             suppressWarnings(file.remove(icudtzipfname))
             # download icudt
-            if (download.file(paste(href, fname, sep = ""), icudtzipfname, mode = "wb") !=                 0) {
+            if (download.file(paste(href, fname, sep = ""), icudtzipfname, mode = "wb") != 0) {
                 return("download error")
             }
             if (!file.exists(icudtzipfname))
@@ -219,11 +185,11 @@ stri_download_icudt <- function(inpath, icu_bundle_version)
 stri_install_icudt <- function(check = TRUE, outpath = NULL, inpath = NULL, icu_bundle_version = NULL)
 {
     # As of v1.1.3, this function is no longer exported.
-    # It was deprecated in 0.5-5.
+    # It was deprecated in v0.5-5.
 
     stopifnot(is.logical(check), length(check) == 1, !is.na(check))
     if (check && stri_install_check(TRUE)) {
-        message("icudt is already installed.")
+        message("icudt has already been installed.")
         return(invisible(TRUE))
     }
 
@@ -240,10 +206,10 @@ stri_install_icudt <- function(check = TRUE, outpath = NULL, inpath = NULL, icu_
 
     stopifnot(is.character(outpath), length(outpath) == 1, file.exists(outpath))
 
-    message("decompressing icudt archive ", icudtzipfname, " to: ", outpath)
+    message("decompressing icudt ", icudtzipfname, " to: ", outpath)
     res <- unzip(icudtzipfname, exdir = outpath, overwrite = TRUE)
     if (!is.character(res) || length(res) <= 0) {
-        message("error decompressing icudt archive")
+        message("error decompressing icudt")
         return(invisible(FALSE))
     }
 

README.md

@@ -30,7 +30,7 @@ Available features include:
 * pattern searching (e.g., with Java-like regular expressions),
 * collation and sorting,
 * random string generation,
-* case mapping,
+* case mapping and folding,
 * string transliteration,
 * Unicode normalisation,
 * date-time formatting and parsing,
@@ -44,9 +44,10 @@ and many more.
 **Authors and Contributors**: [Marek Gagolewski](https://www.gagolewski.com/),
 with contributions from Bartłomiej Tartanus and many others.
 
-The package's API was inspired by
-Hadley Wickham's [stringr](https://stringr.tidyverse.org/)
-package (and since 2015 *stringr* powered by *stringi*).
+The package's API was inspired by that of the early (pre-tidyverse; v0.6.2)
+version of Hadley Wickham's
+[stringr](https://cran.r-project.org/web/packages/stringr/)
+package (and since the 2015 v1.0.0 *stringr* is powered by *stringi*).
 
 
 
@@ -56,26 +57,26 @@ package (and since 2015 *stringr* powered by *stringi*).
 
 [How to access the stringi C++ API from within an Rcpp-based R package](https://github.com/gagolews/ExampleRcppStringi)
 
-**System Requirements**: *R >= 2.14*, *ICU4C >= 55* (refer to the
+**System Requirements**: *R >= 3.1*, *ICU4C >= 55* (refer to the
 [INSTALL](https://raw.githubusercontent.com/gagolews/stringi/master/INSTALL)
 file for more details)
 
 **License**: *stringi*'s source code is licensed under the open source
 BSD-3-clause, for more details see the
 [LICENSE](https://raw.githubusercontent.com/gagolews/stringi/master/LICENSE) file.
 
-This *git* repository also contains a custom subset of *ICU4C 55.1*
-and *ICU4C 61.1* source code which is copyrighted by Unicode and others.
+This *git* repository also contains a custom subset of *ICU4C* source code
+which is copyrighted by Unicode, Inc. and others.
 A binary version of the Unicode Character Database is included.
 For more details on copyright holders see the
 [LICENSE](https://raw.githubusercontent.com/gagolews/stringi/master/LICENSE) file.
 The *ICU* project is covered by the
-[ICU license](http://source.icu-project.org/repos/icu/icu/trunk/LICENSE) —
+[Unicode license](https://github.com/unicode-org/icu/blob/main/icu4c/LICENSE) —
 a simple, permissive non-copyleft free software license, compatible with
 the GNU GPL. The *ICU* license
 is [intended](http://userguide.icu-project.org/icufaq#TOC-How-is-the-ICU-licensed-)
-to allow *ICU* to be included both in free software projects
-and in proprietary or commercial products.
+to allow *ICU* to be included in free software projects as well as
+in proprietary or commercial products.
 
 **Changes**: see the
 [NEWS](https://raw.githubusercontent.com/gagolews/stringi/master/NEWS) file.

configure.win

@@ -1,5 +1,5 @@
 # `stringi` configure.win
-# (C) 2015-2018 Marek Gagolewski
+# Copyright (c) 2013-2021, Marek Gagolewski <https://www.gagolewski.com>
 
 # this is an architecture-independent configure.win file
 

devel/sphinx/index.rst

@@ -21,7 +21,7 @@ It gives you a multitude of functions for:
 * pattern searching (e.g., with ICU Java-like regular expressions),
 * collation and sorting,
 * random string generation,
-* case mapping,
+* case mapping and folding,
 * string transliteration,
 * Unicode normalisation,
 * date-time formatting and parsing,
@@ -45,9 +45,10 @@ by calling:
 It has been released under the open source BSD-3-clause
 `license <https://raw.githubusercontent.com/gagolews/stringi/master/LICENSE>`_.
 
-The package's API was inspired by Hadley Wickham's
-`stringr <https://stringr.tidyverse.org/>`_ package
-(and since 2015 `stringr` is powered by `stringi`).
+The package's API was inspired by that of the early (pre-tidyverse; v0.6.2)
+version of Hadley Wickham's
+`stringr <https://cran.r-project.org/web/packages/stringr/>`_
+package (and since the 2015 v1.0.0 `stringr` is powered by `stringi`).
 Moreover, Hadley suggested quite a few new package features.
 The contributions from Bartłomiej Tartanus and
 `many others <https://github.com/gagolews/stringi/graphs/contributors>`_