CRAN release v0.5.2

.Rbuildignore

@@ -7,6 +7,7 @@
 ^src/cwb/doc$
 ^src/cwb/man$
 ^src/cwb/instutils$
+^src/cwb/editline$
 ^_pkgdown\.yml$
 ^docs$
 ^CONDUCT.md$
@@ -15,3 +16,4 @@
 ^include$
 ^CRAN-RELEASE$
 ^\.github$
+^CRAN-SUBMISSION$

.github/workflows/R-CMD-check.yaml

@@ -26,7 +26,7 @@ jobs:
       matrix:
         config:
           - {os: windows-latest, r: 'release'}
-          - {os: windows-2022, r: 'devel'}
+#          - {os: windows-2022, r: 'devel'}
           - {os: macOS-latest, r: 'release'}
           - {os: ubuntu-20.04, r: 'release', rspm: "https://packagemanager.rstudio.com/cran/__linux__/focal/latest"}
           - {os: ubuntu-20.04, r: 'devel', rspm: "https://packagemanager.rstudio.com/cran/__linux__/focal/latest"}
@@ -46,18 +46,18 @@ jobs:
           http-user-agent: ${{ matrix.config.http-user-agent }}
           use-public-rspm: true
 
-      - name: Setup R (Windows UCRT)
-        if: matrix.config.os == 'windows-2022'
-        uses: kalibera/ucrt3/actions/r-install@main
+#      - name: Setup R (Windows UCRT)
+#        if: matrix.config.os == 'windows-2022'
+#        uses: kalibera/ucrt3/actions/r-install@main
 
-      - name: Install UCRT toolchain
-        if: matrix.config.os == 'windows-2022'
-        uses: kalibera/ucrt3/actions/toolchain-install@main
-        with:
-          # base ... toolchain has the compilers and libraries to build R and recommended packages
-          # full ... additional libraries to build CRAN packages
-          # none ... no toolchain is needed (no native code) 
-         toolchain-type: full      
+#      - name: Install UCRT toolchain
+#        if: matrix.config.os == 'windows-2022'
+#        uses: kalibera/ucrt3/actions/toolchain-install@main
+#        with:
+#          # base ... toolchain has the compilers and libraries to build R and recommended packages
+#          # full ... additional libraries to build CRAN packages
+#          # none ... no toolchain is needed (no native code) 
+#         toolchain-type: full      
 
       - uses: r-lib/actions/setup-pandoc@v2
 
@@ -107,12 +107,19 @@ jobs:
           rcmdcheck::rcmdcheck(args = c("--no-manual", "--as-cran"), error_on = "warning", check_dir = "check")
         shell: Rscript {0}
 
-      - name: Check package (UCRT)
-        if: matrix.config.os == 'windows-2022'
-        env:
-          _R_INSTALL_TIME_PATCHES_: no
-          TZ: UTC
-        uses: kalibera/R-actions/pkg-check@master
+#      - name: Whereabouts of gcc and g++
+#        if: matrix.config.os == 'windows-2022'
+#        run: |
+#          echo %PATH%
+#          where gcc
+#          where g++
+
+#      - name: Check package (UCRT)
+#        if: matrix.config.os == 'windows-2022'
+#        env:
+#          _R_INSTALL_TIME_PATCHES_: no
+#          TZ: UTC
+#        uses: kalibera/R-actions/pkg-check@master
 
       - name: Build Windows binary package
         if: matrix.os == 'windows-latest'

DESCRIPTION

@@ -1,8 +1,8 @@
 Package: RcppCWB
 Type: Package
 Title: 'Rcpp' Bindings for the 'Corpus Workbench' ('CWB')
-Version: 0.5.0
-Date: 2022-01-31
+Version: 0.5.2
+Date: 2022-03-28
 Author: Andreas Blaette [aut, cre],
   Bernard Desgraupes [aut],
   Sylvain Loiseau [aut],
@@ -40,21 +40,21 @@ License: GPL-3
 Encoding: UTF-8
 Copyright: For the copyrights for the 'Corpus Workbench' (CWB) and acknowledgement of authorship, see file COPYRIGHTS.
 NeedsCompilation: yes
-SystemRequirements: GNU make, ncurses, pcre (>= 7 < 10), GLib (>= 2.0.0). On Windows, no prior 
-  installations are necessary, as pre-built (i.e. cross-compiled) binaries of required libraries are 
-  downloaded from a GitHub repository (<https://github.com/PolMine/libcl>) during installation. On 
-  macOS, static libraries of Glib are downloaded (<https://github.com/PolMine/libglib>) if Glib is 
-  not present.
+SystemRequirements: GNU make, pcre (>= 7 < 10), GLib (>= 2.0.0). On Windows, no prior installations are necessary,
+  as pre-built (i.e. cross-compiled) binaries of required libraries are downloaded from a GitHub repository 
+  (<https://github.com/PolMine/libcl>) during installation. On macOS, static libraries of Glib are downloaded
+  (<https://github.com/PolMine/libglib>) if Glib is not present.
 Imports:
-    Rcpp (>= 1.0.7)
+    Rcpp (>= 1.0.7),
+    fs
 Suggests:
     knitr,
     testthat
 LinkingTo: Rcpp
 Biarch: true
 URL: https://github.com/PolMine/RcppCWB
 BugReports: https://github.com/PolMine/RcppCWB/issues
-RoxygenNote: 7.1.1
+RoxygenNote: 7.1.2
 Roxygen: list(markdown = TRUE)
 Collate:
   'RcppCWB_package.R'

NAMESPACE

@@ -2,28 +2,38 @@
 
 export(check_corpus)
 export(check_cpos)
-export(check_cqp_query)
 export(check_id)
 export(check_p_attribute)
 export(check_pkg_registry_files)
+export(check_query)
 export(check_region_matrix)
 export(check_registry)
 export(check_s_attribute)
 export(check_strucs)
 export(cl_charset_name)
 export(cl_delete_corpus)
+export(cl_find_corpus)
 export(cl_struc_values)
 export(corpus_data_dir)
 export(corpus_is_loaded)
+export(cpos_to_id)
+export(cpos_to_lbound)
+export(cpos_to_rbound)
+export(cpos_to_str)
+export(cpos_to_struc)
+export(cqp_drop_subcorpus)
 export(cqp_dump_subcorpus)
 export(cqp_get_registry)
 export(cqp_initialize)
 export(cqp_is_initialized)
 export(cqp_list_corpora)
 export(cqp_list_subcorpora)
+export(cqp_load_corpus)
 export(cqp_query)
 export(cqp_reset_registry)
 export(cqp_subcorpus_size)
+export(cqp_verbosity)
+export(cwb_charsets)
 export(cwb_compress_rdx)
 export(cwb_encode)
 export(cwb_huffcode)
@@ -34,12 +44,29 @@ export(get_count_vector)
 export(get_pkg_registry)
 export(get_region_matrix)
 export(get_tmp_registry)
+export(id_to_cpos)
+export(id_to_freq)
+export(p_attr)
+export(p_attr_lexicon_size)
+export(p_attr_size)
+export(ranges_to_cpos)
+export(regex_to_id)
+export(region_matrix_context)
 export(region_matrix_to_count_matrix)
 export(region_matrix_to_ids)
+export(s_attr)
 export(s_attr_is_descendent)
 export(s_attr_is_sibling)
+export(s_attr_relationship)
+export(s_attr_size)
 export(s_attribute_decode)
+export(str_to_id)
+export(struc_to_cpos)
+export(struc_to_str)
 export(use_tmp_registry)
 exportPattern("^[[:alpha:]]+")
 importFrom(Rcpp,evalCpp)
+importFrom(fs,path)
+importFrom(fs,path_expand)
+importFrom(utils,capture.output)
 useDynLib(RcppCWB, .registration = TRUE)

NEWS.md

@@ -1,3 +1,102 @@
+# RcppCWB 0.5.2
+
+* The example for `corpus_data_dir()` dir not work as intended without
+explicitly setting the `registry` argument. Fixed.
+* New functions `corpus_info_file()`, `corpus_full_name()`,
+`corpus_p_attributes()`, `corpus_s_attributes()`, `corpus_properties()` and
+`corpus_property()` to retrieve registry file data.
+* New function `corpus_registry_dir()`.
+* The path to the info file in the registry file of the REUTERS corpus was
+broken. Fixed.
+
+
+# RcppCWB 0.5.1
+
+## New Features
+
+* New auxiliary function `cwb_charsets()` reports the charsets supported by CWB.
+* New functions `cl_load_corpus()` and `cqp_load_corpus()` do what the functions
+suggests.
+* New function `cl_list_corpora()` complements existing function
+`cqp_list_corpora()` for the CL context.
+* New arguments `skip_blank_lines`, `strip_whitespace` and `xml` of
+`cwb_encode()` open configuration options of `cwb_encode()`, overcoming the
+previously hard-coded equivalent to the command-line option "-xsB".(#38)
+* Unexported functions `.cpos_to_id()`, `.cl_find_corpus()` and
+`.cl_new_attribute()` are an entry to passing around pointers, rather than
+re-creating objects whenever switching from R to C.
+* Functions `.s_attr()` and `.p_attr()` return pointers for a s- or
+p-attribute.
+* Functions `cl_*` are now available with pointer as input (e.g. `cpos_to_id()`).
+* The CORPUS_REGISTRY environment variable is not set to the temporary registry,
+to avoid often confusing behavior and collissions whent loading RcppCWB and 
+polmineR at the same time (#13).
+* The `cqp_drop_subcorpus()` function that has been disabled temporarily is
+usable again (#34).
+* `cqp_query()` is now able to process subcorpora.
+* `RcppCWB:::.cqp_subcropus()` will construct a subcorpus from a region matrix.
+* The `check_corpus()` does not re-set the registry directory and more, but tries
+to load the checked corpus if it has not yet been loaded.
+* A new function `s_attr_relationship()` will detect whether two s-attributes are
+siblings, or in a descendent or ancestor relationship.
+* Functions `cwb_encode()`, `cwb_huffcode()`, `cwb_makeall()` and
+`cwb_compress_rdx()` now have an argument `quietly` to control display of output
+messages. `cwb_encode()` has an argument `verbose` to control whether counter on
+the number of tokens processed is dislpayed.
+
+
+## Minor improvements
+
+* Difficulties of `cwb_encode()` to digest variations of path statements between
+macOS and Windows are addressed using a reliable normalization of paths with
+`fs::path()` (#48).
+* Argument `encoding` is checked for the validity of the encoding passed in
+(#34).
+* A patch introducing a sanity check omits 'stringop-overflow'  compiler warning
+thrown by file cl/cdaccess.c on Windows (#45).
+* An update of Xcode command line developer tools includes flex 2.6.4
+Apple(flex-34), and this is the version used not, resulting and extensive code
+changes in cl/lex.creg.c and cqp/lex.yy.c, yet without causing new errors or
+changing the functionality.
+* `check_cpos()` issues a warning if argument `cpos` is `NULL` (#21).
+* Functions `cl_cpos2id()`, `cl_cpos2lbound()`, `cl_cpos2rbound()`,
+`cl_cpos2str()` and `cl_cpo2struc()` will return an empty, zero-length integer
+vector if argument `cpos` is `NULL` (#21).
+* Warnings issued by `check_corpus()` (used internally by many functions)
+resulted from slightly differing representations of otherwise identical 
+paths. Using `fs::path()` for path for normalization internally will omit
+misleading warning messages.
+* `cqp_get_registry()` will now return a `fs::path` object, as a safeguard for
+a consistent normalization of paths.
+* Function `cl_delete_corpus()` will now (visibly) return a `logial` value.
+* The check for the availability of ncurses is omitted in the configure file
+and the editline subdirectory of src/cwb is included in .Rbuildignore to 
+minimize the size of the tarball. The ncurses library is a dependency of 
+editline, but editline is not built in the context of this package (#26).
+* `cqp_load_corpus()` will return `FALSE` if corpus has not been loaded
+successfully.
+* Disaggregated `wrappers.cpp` into `cl.cpp`, `cqp.cpp` and `utils.cpp`, so that
+the code is organized more coherently corresponding to the different logics.
+* Function `check_cqp_query()` renamed to `check_query()` to avoid a conflict
+with a function defined in the polmineR package.
+* `cqp_list_subcorpora()` returns a `character` vector. Previously, we just had
+obscure printed messages.
+* `s_attribute_decode()` will not break if s-attribute has no values (#54).
+* Functions `cl_struc2str()` and `cl_struc2cpos()` may now include negative
+values, the vectors returned will have `NA` values at respective positions. The
+check against negative values in `check_strucs` is dropped accordingly.
+
+
+## Bux fixes
+
+* The `cwb_encode()` function did not declare structural attributes in the
+registry and mistakenly channeled output for the file to the terminal (#49).
+Fixed.
+* Re-running `cwb_encode()` did not reset global variables, which resulted in a
+set of errors. Solved. (#51)
+
+
+
 # RcppCWB 0.5.0
 
 ## New Features

R/RcppCWB_package.R

@@ -20,7 +20,8 @@
 #'   The CWB is a classical tool which has inspired a set of developments. A
 #'   persisting advantage of the CWB is its mature, open source code base that
 #'   is actively maintained by a community of developers. It is used as a robust
-#'   and efficient backend for widely used tools such as TXM or CQPweb
+#'   and efficient backend for widely used tools such as
+#'   TXM(\url{https://txm.gitpages.huma-num.fr/textometrie/}) or CQPweb
 #'   (\url{https://cwb.sourceforge.io/cqpweb.php}). Its uncompromising C
 #'   implementation guarantees speed and makes it well suited to be integrated
 #'   with R at the same time.
@@ -117,27 +118,22 @@
 #' # functions of the corpus library (starting with cl) expose the low-level
 #' # access to the CWB corpus library (CL)
 #' 
-#' # registry <- if (!check_pkg_registry_files()) use_tmp_registry() else get_pkg_registry()
-#' registry <- use_tmp_registry()
-#' print(registry)
-#' ids <- cl_cpos2id("REUTERS", cpos = 1:20, p_attribute = "word", registry = registry)
-#' tokens <- cl_id2str("REUTERS", id = ids, p_attribute = "word", registry = registry)
+#' ids <- cl_cpos2id("REUTERS", cpos = 1:20, p_attribute = "word", registry = get_tmp_registry())
+#' tokens <- cl_id2str("REUTERS", id = ids, p_attribute = "word", registry = get_tmp_registry())
 #' print(paste(tokens, collapse = " "))
 #' 
 #' # To use the corpus query processor (CQP) and its syntax, it is necessary first
 #' # to initialize CQP (example: get concordances of 'oil')
 #' 
-#' cqp_initialize(registry)
 #' cqp_query("REUTERS", query = '[]{5} "oil" []{5}')
 #' cpos_matrix <- cqp_dump_subcorpus("REUTERS")
 #' concordances_oil <- apply(
 #'   cpos_matrix, 1,
 #'   function(row){
-#'     ids <- cl_cpos2id("REUTERS", p_attribute = "word", cpos = row[1]:row[2])
-#'     tokens <- cl_id2str("REUTERS", p_attribute = "word", id = ids)
+#'     ids <- cl_cpos2id("REUTERS", p_attribute = "word", cpos = row[1]:row[2], get_tmp_registry())
+#'     tokens <- cl_id2str("REUTERS", p_attribute = "word", id = ids, get_tmp_registry())
 #'     paste(tokens, collapse = " ")
 #'   }
 #'  )
-#' 
 NULL