Bump to 0.4.0: worksheet & dashboard intelligence

.Rbuildignore

@@ -19,3 +19,5 @@ cran_examples_sanitizer.R
 ^check/
 ^twbparser.Rcheck$
 ^\.Rd2pdf20656$
+^\.git$
+^\.claude$

DESCRIPTION

@@ -1,6 +1,6 @@
 Package: twbparser
 Title: Parse 'Tableau' Workbooks into Functional Data
-Version: 0.3.1
+Version: 0.4.0
 Authors@R: 
     person("George", "Arthur", , "prigasgenthian48@gmail.com", role = c("aut", "cre"),
            comment = c(ORCID = "0000-0002-1975-1459"))
@@ -46,4 +46,4 @@ Encoding: UTF-8
 Language: en-US
 LazyData: false
 Roxygen: list(markdown = TRUE, r6 = FALSE)
-RoxygenNote: 7.3.2
+RoxygenNote: 7.3.3

NAMESPACE

@@ -23,14 +23,21 @@ export(tbs_publish_info)
 export(twb_charts)
 export(twb_colors)
 export(twb_custom_sql)
+export(twb_dashboard_actions)
 export(twb_dashboard_filters)
+export(twb_dashboard_layout)
+export(twb_dashboard_sheets)
 export(twb_dashboard_summary)
 export(twb_dashboards)
 export(twb_initial_sql)
 export(twb_page_composition)
 export(twb_pages)
 export(twb_pages_summary)
 export(twb_published_refs)
+export(twb_sheet_axes)
+export(twb_sheet_filters)
+export(twb_sheet_shelves)
+export(twb_sheet_sorts)
 export(twbx_extract_files)
 export(twbx_list)
 export(validate_relationships)
@@ -63,6 +70,7 @@ importFrom(igraph,graph_from_data_frame)
 importFrom(igraph,layout_with_fr)
 importFrom(igraph,make_empty_graph)
 importFrom(purrr,map)
+importFrom(purrr,map_chr)
 importFrom(purrr,map_dfr)
 importFrom(rlang,.data)
 importFrom(stringr,str_detect)
@@ -82,7 +90,9 @@ importFrom(withr,with_seed)
 importFrom(xml2,read_xml)
 importFrom(xml2,xml_attr)
 importFrom(xml2,xml_attrs)
+importFrom(xml2,xml_children)
 importFrom(xml2,xml_find_all)
 importFrom(xml2,xml_find_first)
 importFrom(xml2,xml_name)
 importFrom(xml2,xml_parent)
+importFrom(xml2,xml_text)

NEWS.md

@@ -1,3 +1,55 @@
+# twbparser 0.4.0
+
+## New features
+
+### Per-worksheet intelligence
+
+* `twb_sheet_shelves()` — extract every field placed on rows, cols, or an
+  encoding shelf (color, size, label, detail, tooltip) for one or all worksheets.
+* `twb_sheet_filters()` — extract worksheet-level filters including categorical
+  member lists, range min/max, and include/exclude mode.
+* `twb_sheet_axes()` — extract per-axis configuration: reversed, include-zero,
+  and scale type (linear, log, …).
+* `twb_sheet_sorts()` — extract sort directives with sort direction and method
+  (field aggregate, alphabetic, manual, data-source order).
+* All four are exposed on `TwbParser` as `get_sheet_shelves()`,
+  `get_sheet_filters()`, `get_sheet_axes()`, `get_sheet_sorts()` and as
+  no-parens active bindings (`parser$sheet_shelves`, etc.).
+
+### Per-dashboard intelligence
+
+* `twb_dashboard_sheets()` — list every worksheet placed on a dashboard with
+  zone id and pixel position (x, y, w, h).
+* `twb_dashboard_layout()` — full zone tree including parent zone id, component
+  type (worksheet / filter / container / …), layout type (tiled / floating),
+  and pixel bounds.
+* `twb_dashboard_actions()` — extract filter and URL actions with source and
+  target sheets, run-on trigger type, and URL value.
+* All three are exposed on `TwbParser` as `get_dashboard_sheets()`,
+  `get_dashboard_layout()`, `get_dashboard_actions()` and as active bindings.
+
+## Bug fixes
+
+* `plot_relationship_graph()`: fixed edge direction — the `from` vertex was
+  incorrectly built from `right_field` instead of `left_field`.
+* `plot_source_join_graph()`: fixed reference to non-existent columns
+  `left_source` / `right_source`; replaced with `left_table` / `right_table`.
+* `infer_implicit_relationships()`: added deduplication before the field-name
+  self-join to prevent Cartesian explosion when many tables share a field name;
+  added `relationship = "many-to-many"` to suppress the dplyr 1.1+ warning.
+* Fixed `integer_` typo in `insights.R` that caused errors when parsing
+  dashboard zone dimensions.
+
+## Internal
+
+* Canonical `.twb_clean_table()` and `.twb_clean_field()` helpers added to
+  `utils.R`, replacing four independent copies scattered across `fields.R`,
+  `calculated_fields.R`, `relationships.R`, `joins.R`, and `dependency_graph.R`.
+* `.twb_clean_field()` now correctly strips Tableau column-instance prefixes
+  (`none:Category:nk` → `Category`) and returns unnamed vectors.
+
+---
+
 # twbparser 0.3.1
 
 * Remove use of `unlockBinding()` in internal TwbParser active-binding helpers.

R/active-bindings.R

@@ -91,6 +91,15 @@ twb_install_active_properties <- function(x, cache = TRUE) {
   rebind("twbx_extracts_tbl",  wrap_cache("twbx_extracts_tbl",  function() x$get_twbx_extracts()))
   rebind("twbx_images_tbl",    wrap_cache("twbx_images_tbl",    function() x$get_twbx_images()))
 
+  ## Phase 2/3: sheet & dashboard intelligence
+  rebind("sheet_shelves",      wrap_cache("sheet_shelves",      function() x$get_sheet_shelves()))
+  rebind("sheet_filters",      wrap_cache("sheet_filters",      function() x$get_sheet_filters()))
+  rebind("sheet_axes",         wrap_cache("sheet_axes",         function() x$get_sheet_axes()))
+  rebind("sheet_sorts",        wrap_cache("sheet_sorts",        function() x$get_sheet_sorts()))
+  rebind("dashboard_sheets",   wrap_cache("dashboard_sheets",   function() x$get_dashboard_sheets()))
+  rebind("dashboard_layout",   wrap_cache("dashboard_layout",   function() x$get_dashboard_layout()))
+  rebind("dashboard_actions",  wrap_cache("dashboard_actions",  function() x$get_dashboard_actions()))
+
   ## Validation snapshot (read-only)
   rebind(
     "validation",

R/calculated_fields.R

@@ -3,18 +3,6 @@
   ifelse(is.na(x), NA_character_, gsub("\\[|\\]", "", x))
 }
 
-#' @keywords internal
-.clean_table <- function(x) {
-  if (is.null(x) || is.na(x)) {
-    return(NA_character_)
-  }
-  x <- gsub("^\\[.*?\\]\\.", "", x) # drop [Extract]. / [Connection].
-  x <- gsub("\\[|\\]", "", x) # strip [ ]
-  x <- sub("_[0-9A-Fa-f]{32}$", "", x) # drop 32-hex suffix
-  x <- trimws(x)
-  if (!nzchar(x)) NA_character_ else x
-}
-
 #' Extract calculated fields from a TWB
 #'
 #' Finds columns that contain \code{<calculation>}nodes and returns metadata and
@@ -103,7 +91,7 @@ extract_calculated_fields <- function(xml_doc, include_parameters = FALSE) {
         calc_class            = calc_class,
         is_table_calc         = is_tbl,
         table                 = raw_tbl,
-        table_clean           = .clean_table(raw_tbl)
+        table_clean           = .twb_clean_table(raw_tbl)
       )
     })
   }) |>
@@ -181,7 +169,7 @@ extract_parameters <- function(xml_doc) {
         current_value         = cur_val,
         is_parameter          = TRUE,
         table                 = raw_tbl,
-        table_clean           = .clean_table(raw_tbl)
+        table_clean           = .twb_clean_table(raw_tbl)
       )
     })
   }) |>
@@ -251,7 +239,7 @@ extract_raw_fields <- function(xml_doc) {
         is_hidden             = attr_safe_get(a, "hidden", "false") %in% c("true", "1"),
         is_parameter          = FALSE,
         table                 = raw_tbl,
-        table_clean           = .clean_table(raw_tbl)
+        table_clean           = .twb_clean_table(raw_tbl)
       )
     })
   }) |>