Fix some typos (most found by codespell)

Signed-off-by: Stefan Weil <[email protected]>

Author: stweil

doc/Doxyfile

@@ -1656,7 +1656,7 @@ EXTRA_PACKAGES         =
 # Note: Only use a user-defined header if you know what you are doing! The
 # following commands have a special meaning inside the header: $title,
 # $datetime, $date, $doxygenversion, $projectname, $projectnumber,
-# $projectbrief, $projectlogo. Doxygen will replace $title with the empy string,
+# $projectbrief, $projectlogo. Doxygen will replace $title with the empty string,
 # for the replacement values of the other commands the user is referred to
 # HTML_HEADER.
 # This tag requires that the tag GENERATE_LATEX is set to YES.

java/com/google/scrollview/events/SVEventType.java

@@ -20,7 +20,7 @@
 public enum SVEventType {
   SVET_DESTROY, // Window has been destroyed by user.
   SVET_EXIT, // User has destroyed the last window by clicking on the 'X'
-  SVET_CLICK, // Any button pressed thats not a popup trigger. 
+  SVET_CLICK, // Any button pressed that is not a popup trigger.
   SVET_SELECTION, // Left button selection.
   SVET_INPUT, // Any kind of input
   SVET_MOUSE, // The mouse has moved with a button pressed.

java/com/google/scrollview/ui/SVWindow.java

@@ -254,7 +254,7 @@ public SVWindow(String name, int hash, int posX, int posY, int sizeX,
     layer = canvas.getLayer();
     canvas.setBackground(Color.BLACK);
 
-    // Disable anitaliasing to make the lines more visible.
+    // Disable antialiasing to make the lines more visible.
     canvas.setDefaultRenderQuality(PPaintContext.LOW_QUALITY_RENDERING);
 
     setLayout(new BorderLayout());

src/ccmain/equationdetect.cpp

@@ -1029,7 +1029,7 @@ EquationDetect::IndentType EquationDetect::IsIndented(ColPartition* part) {
   const int kYGapTh = static_cast<int>(roundf(0.5 * resolution_));
 
   // Here we use a simple approximation algorithm: from the center of part, We
-  // perform the radius search, and check if we can find a neighboring parition
+  // perform the radius search, and check if we can find a neighboring partition
   // that locates on the top/bottom left of part.
   search.StartRadSearch((part_box.left() + part_box.right()) / 2,
       (part_box.top() + part_box.bottom()) / 2, kRadiusTh);

src/ccmain/equationdetect.h

@@ -136,10 +136,10 @@ class EquationDetect : public EquationDetectBase {
   bool CheckSeedFgDensity(const float density_th, ColPartition* part);
 
   // A light version of SplitCPHor: instead of really doing the part split, we
-  // simply compute the union bounding box of each splitted part.
+  // simply compute the union bounding box of each split part.
   void SplitCPHorLite(ColPartition* part, GenericVector<TBOX>* splitted_boxes);
 
-  // Split the part (horizontally), and save the splitted result into
+  // Split the part (horizontally), and save the split result into
   // parts_splitted. Note that it is caller's responsibility to release the
   // memory owns by parts_splitted. On the other hand, the part is unchanged
   // during this process and still owns the blobs, so do NOT call DeleteBoxes
@@ -159,7 +159,7 @@ class EquationDetect : public EquationDetectBase {
   // Identify inline partitions from cp_seeds_, and re-label them.
   void IdentifyInlineParts();
 
-  // Comute the super bounding box for all colpartitions inside part_grid_.
+  // Compute the super bounding box for all colpartitions inside part_grid_.
   void ComputeCPsSuperBBox();
 
   // Identify inline partitions from cp_seeds_ using the horizontal search.
@@ -244,7 +244,7 @@ class EquationDetect : public EquationDetectBase {
   // ColPartition object.
   void PrintSpecialBlobsDensity(const ColPartition* part) const;
 
-  // The tesseract engine intialized from equation training data.
+  // The tesseract engine initialized from equation training data.
   Tesseract equ_tesseract_;
 
   // The tesseract engine used for OCR. This pointer is passed in by the caller,

src/ccmain/fixspace.cpp

@@ -226,7 +226,7 @@ void Tesseract::match_current_words(WERD_RES_LIST &words, ROW *row,
  * The solution is to NOT COUNT the score of any word which has a digit at one
  * end and a "1Il" as the character the other side of the space.
  *
- * Conversly, any character next to a "1" within a word is counted as a positive
+ * Conversely, any character next to a "1" within a word is counted as a positive
  * score. Thus "561 63" would score 4 (3 chars in a numeric word plus 1 side of
  * the "1" joined).  "56163" would score 7 - all chars in a numeric word + 2
  * sides of a "1" joined.

src/ccmain/paragraphs.cpp

@@ -1641,7 +1641,7 @@ bool FirstWordWouldHaveFit(const RowScratchRegisters &before,
 
 // Return whether the first word on the after line can fit in the space at
 // the end of the before line (not knowing which way the text goes) in a left
-// or right alignemnt.
+// or right alignment.
 bool FirstWordWouldHaveFit(const RowScratchRegisters &before,
                            const RowScratchRegisters &after) {
   if (before.ri_->num_words == 0 || after.ri_->num_words == 0)

src/ccmain/superscript.cpp

@@ -91,7 +91,7 @@ void YOutlierPieces(WERD_RES *word, int rebuilt_blob_index,
 /**
  * Attempt to split off any high (or low) bits at the ends of the word with poor
  * certainty and recognize them separately.  If the certainty gets much better
- * and other sanity checks pass, acccept.
+ * and other sanity checks pass, accept.
  *
  * This superscript fix is meant to be called in the second pass of recognition
  * when we have tried once and already have a preliminary answer for word.

src/ccmain/tessedit.cpp

@@ -112,7 +112,7 @@ bool Tesseract::init_tesseract_lang_data(
     return false;
   }
   if (oem == OEM_DEFAULT) {
-    // Set the engine mode from availability, which can then be overidden by
+    // Set the engine mode from availability, which can then be overridden by
     // the config file when we read it below.
     if (!mgr->IsLSTMAvailable()) {
       tessedit_ocr_engine_mode.set_value(OEM_TESSERACT_ONLY);

src/ccmain/tesseractclass.cpp

@@ -334,7 +334,7 @@ Tesseract::Tesseract()
       double_MEMBER(fixsp_small_outlines_size, 0.28, "Small if lt xht x this",
                     this->params()),
       BOOL_MEMBER(tessedit_prefer_joined_punct, false,
-                  "Reward punctation joins", this->params()),
+                  "Reward punctuation joins", this->params()),
       INT_MEMBER(fixsp_done_mode, 1, "What constitues done for spacing",
                  this->params()),
       INT_MEMBER(debug_fix_space_level, 0, "Contextual fixspace debug",
@@ -621,7 +621,7 @@ void Tesseract::PrepareForPageseg() {
     sub_langs_[i]->pix_binary_ = pixClone(pix_binary());
   }
   // Perform shiro-rekha (top-line) splitting and replace the current image by
-  // the newly splitted image.
+  // the newly split image.
   splitter_.set_orig_pix(pix_binary());
   splitter_.set_pageseg_split_strategy(max_pageseg_strategy);
   if (splitter_.Split(true, &pixa_debug_)) {

src/ccmain/tesseractclass.h

@@ -993,7 +993,7 @@ class Tesseract : public Wordrec {
   INT_VAR_H(fixsp_non_noise_limit, 1,
             "How many non-noise blbs either side?");
   double_VAR_H(fixsp_small_outlines_size, 0.28, "Small if lt xht x this");
-  BOOL_VAR_H(tessedit_prefer_joined_punct, false, "Reward punctation joins");
+  BOOL_VAR_H(tessedit_prefer_joined_punct, false, "Reward punctuation joins");
   INT_VAR_H(fixsp_done_mode, 1, "What constitues done for spacing");
   INT_VAR_H(debug_fix_space_level, 0, "Contextual fixspace debug");
   STRING_VAR_H(numeric_punctuation, ".,",

src/ccstruct/crakedge.h

@@ -1,6 +1,6 @@
 /**********************************************************************
  * File:						crakedge.h      (Formerly: crkedge.h)
- * Description: Sturctures for the Crack following edge detector.
+ * Description: Structures for the Crack following edge detector.
  * Author:					Ray Smith
  * Created:					Fri Mar 22 16:06:38 GMT 1991
  *

src/ccstruct/linlsq.cpp

@@ -152,7 +152,7 @@ double LLSQ::rms(double m,  double c) const {          // get error
  **********************************************************************/
 
 double LLSQ::pearson() const {  // get correlation
-  double r = 0.0;                  // Correlation is 0 if insufficent data.
+  double r = 0.0;                  // Correlation is 0 if insufficient data.
 
   double covar = covariance();
   if (covar != 0.0) {

src/ccstruct/matrix.h

@@ -5,7 +5,7 @@
  * Author:       Ray Smith
  * TODO(rays) Separate from ratings matrix, which it also contains:
  *
- * Descrition:   Ratings matrix class (specialization of banded matrix).
+ * Description:  Ratings matrix class (specialization of banded matrix).
  *               Segmentation search matrix of lists of BLOB_CHOICE.
  * Author:       Mark Seaman, OCR Technology
  * Created:      Wed May 16 13:22:06 1990

src/ccutil/elst.h

@@ -67,7 +67,7 @@ The implementation of lists is very careful about space and speed overheads.
 This is why many embedded lists are provided. The same concerns mean that
 in-line type coercion is done, rather than use virtual functions.  This is
 cumbersome in that each data type to be listed requires its own iterator and
-list class - though macros can gererate these.  It also prevents heterogeneous
+list class - though macros can generate these.  It also prevents heterogeneous
 lists.
 **********************************************************************/
 

src/ccutil/genericvector.h

@@ -343,7 +343,7 @@ class GenericVector {
   // Init the object, allocating size memory.
   void init(int size);
 
-  // We are assuming that the object generally placed in thie
+  // We are assuming that the object generally placed in the
   // vector are small enough that for efficiency it makes sense
   // to start with a larger initial size.
   static const int kDefaultVectorSize = 4;

src/ccutil/qrsequence.h

@@ -31,7 +31,7 @@
 
 class QRSequenceGenerator {
  public:
-  // Object is initalized with the size of the output range.
+  // Object is initialized with the size of the output range.
   explicit QRSequenceGenerator(int N) : N_(N), next_num_(0) {
     num_bits_ = static_cast<int>(ceil(log(static_cast<double>(N)) / log(2.0)));
   }

src/ccutil/unicharcompress.h

@@ -113,7 +113,7 @@ class RecodedCharID {
 // 2 (Indic): Instead of thousands of codes with one for each grapheme, re-code
 //            as the unicode sequence (but coded in a more compact space).
 // 3 (the rest): Eliminate multi-path problems with ligatures and fold confusing
-//               and not significantly distinct shapes (quotes) togther, ie
+//               and not significantly distinct shapes (quotes) together, ie
 //               represent the fi ligature as the f-i pair, and fold u+2019 and
 //               friends all onto ascii single '
 // 4 The null character and mapping to target activations:

src/ccutil/unicharset.cpp

@@ -138,7 +138,7 @@ void UNICHARSET::UNICHAR_PROPERTIES::SetRangesEmpty() {
 }
 
 // Returns true if any of the top/bottom/width/bearing/advance ranges/stats
-// is emtpy.
+// is empty.
 bool UNICHARSET::UNICHAR_PROPERTIES::AnyRangeEmpty() const {
   return width == 0.0f || advance == 0.0f;
 }

src/ccutil/unicharset.h

@@ -913,7 +913,7 @@ class UNICHARSET {
     // Sets all ranges to empty. Used before expanding with font-based data.
     void SetRangesEmpty();
     // Returns true if any of the top/bottom/width/bearing/advance ranges/stats
-    // is emtpy.
+    // is empty.
     bool AnyRangeEmpty() const;
     // Expands the ranges with the ranges from the src properties.
     void ExpandRangesFrom(const UNICHAR_PROPERTIES& src);

src/classify/cluster.cpp

@@ -2437,7 +2437,7 @@ FLOAT64 ChiArea(CHISTRUCT *ChiParams, FLOAT64 x) {
 /**
  * This routine looks at all samples in the specified cluster.
  * It computes a running estimate of the percentage of the
- * charaters which have more than 1 sample in the cluster.
+ * characters which have more than 1 sample in the cluster.
  * When this percentage exceeds MaxIllegal, TRUE is returned.
  * Otherwise FALSE is returned.  The CharID
  * fields must contain integers which identify the training

src/classify/intmatcher.cpp

@@ -718,7 +718,7 @@ void ScratchEvidence::ClearFeatureEvidence(const INT_CLASS class_template) {
 }
 
 /**
- * Print debugging information for Configuations
+ * Print debugging information for Configurations
  * @return none
  * @note Exceptions: none
  * @note History: Wed Feb 27 14:12:28 MST 1991, RWM, Created.
@@ -741,7 +741,7 @@ void IMDebugConfiguration(int FeatureNum,
 }
 
 /**
- * Print debugging information for Configuations
+ * Print debugging information for Configurations
  * @return none
  * @note Exceptions: none
  * @note History: Wed Feb 27 14:12:28 MST 1991, RWM, Created.
@@ -914,7 +914,7 @@ int IntegerMatcher::UpdateTablesForFeature(
 }
 
 /**
- * Print debugging information for Configuations
+ * Print debugging information for Configurations
  * @return none
  * @note Exceptions: none
  * @note History: Wed Feb 27 14:12:28 MST 1991, RWM, Created.

src/cutil/cutil.cpp

@@ -50,7 +50,7 @@ Import original HP distribution
  * long_rand
  *
  *  Return a long random number whose value is less than limit.  Do this
- *  by calling the standard cheepo random number generator and reseting
+ *  by calling the standard cheepo random number generator and resetting
  *  it pretty often.
  **********************************************************************/
 long long_rand(long limit) {

src/cutil/oldlist.h

@@ -65,7 +65,7 @@
  * sort              - Use quick sort to construct a new list.
  * transform         - Creates a new list by transforming each of the nodes.
  *
- * TRANFORMS:             (Note: These functions all modify the input list.)
+ * TRANSFORMS:             (Note: These functions all modify the input list.)
  * ----------
  * join              - Concatenates list 1 and list 2.
  * delete_d          - Removes the requested elements from the list.

src/dict/dawg.h

@@ -285,7 +285,7 @@ class Dawg {
   void init(int unicharset_size);
 
   /// Matches all of the words that are represented by this string.
-  /// If wilcard is set to something other than INVALID_UNICHAR_ID,
+  /// If wildcard is set to something other than INVALID_UNICHAR_ID,
   /// the *'s in this string are interpreted as wildcards.
   /// WERD_CHOICE param is not passed by const so that wildcard searches
   /// can modify it and work without having to copy WERD_CHOICEs.

src/dict/dict.cpp

@@ -688,7 +688,7 @@ void Dict::adjust_word(WERD_CHOICE *word,
         // leave the factor alone - all good!
         break;
     }
-    // TODO(eger): if nonword is true, but there is a "core" thats' a dict
+    // TODO(eger): if nonword is true, but there is a "core" that is a dict
     // word, negate nonword status.
   } else {
     if (debug) {
@@ -753,7 +753,7 @@ int Dict::valid_word(const WERD_CHOICE &word, bool numbers_ok) const {
   init_active_dawgs(&(active_dawgs[0]), false);
   DawgArgs dawg_args(&(active_dawgs[0]), &(active_dawgs[1]), NO_PERM);
   int last_index = word_ptr->length() - 1;
-  // Call leter_is_okay for each letter in the word.
+  // Call letter_is_okay for each letter in the word.
   for (int i = hyphen_base_size(); i <= last_index; ++i) {
     if (!((this->*letter_is_okay_)(&dawg_args, word_ptr->unichar_id(i),
                                    i == last_index))) break;

src/dict/dict.h

@@ -340,7 +340,7 @@ class Dict {
    * For word_index > 0 the corresponding state (active_dawgs and punc position)
    * can be obtained from dawg_args->updated_dawgs passed to
    * def_letter_is_okay for word_index-1.
-   * Note: the function assumes that active_dawgs, nd updated_dawgs
+   * Note: the function assumes that active_dawgs, and updated_dawgs
    * member variables of dawg_args are not nullptr.
    *
    * Output:
@@ -532,7 +532,7 @@ class Dict {
   Trie *pending_words_;
   /// The following pointers are only cached for convenience.
   /// The dawgs will be deleted when dawgs_ vector is destroyed.
-  // bigram_dawg_ points to a dawg of two-word bigrams which always supercede if
+  // bigram_dawg_ points to a dawg of two-word bigrams which always supersede if
   // any of them are present on the best choices list for a word pair.
   // the bigrams are stored as space-separated words where:
   // (1) leading and trailing punctuation has been removed from each word and

src/dict/stopper.cpp

@@ -178,7 +178,7 @@ bool Dict::NoDangerousAmbig(WERD_CHOICE *best_choice,
       getUnicharAmbigs().replace_ambigs() : getUnicharAmbigs().dang_ambigs();
     if (!replace) {
       // Initialize ambig_blob_choices with lists containing a single
-      // unichar id for the correspoding position in best_choice.
+      // unichar id for the corresponding position in best_choice.
       // best_choice consisting from only the original letters will
       // have a rating of 0.0.
       for (i = 0; i < best_choice->length(); ++i) {

src/dict/trie.h

@@ -402,7 +402,7 @@ class Trie : public Dawg {
                              NODE_MARKER reduced_nodes);
 
   /**
-   * Order num_edges of consequtive EDGE_RECORDS in the given EDGE_VECTOR in
+   * Order num_edges of consecutive EDGE_RECORDS in the given EDGE_VECTOR in
    * increasing order of unichar ids. This function is normally called
    * for all edges in a single node, and since number of edges in each node
    * is usually quite small, selection sort is used.

src/opencl/openclwrapper.cpp

@@ -998,7 +998,7 @@ int OpenclDevice::GeneratBinFromKernelSource( cl_program program, const char * c
         }
     }
 
-    // Release all resouces and memory
+    // Release all resources and memory
     for ( i = 0; i < numDevices; i++ )
     {
       free(binaries[i]);

src/opencl/openclwrapper.h

@@ -241,7 +241,7 @@ class OpenclDevice
 
 /* OpenCL implementations of Morphological operations*/
 
-    //Initialiation of OCL buffers used in Morph operations
+    //Initialization of OCL buffers used in Morph operations
     static int initMorphCLAllocations(l_int32 wpl, l_int32 h, Pix *pixs);
     static void releaseMorphCLBuffers();
 

src/textord/alignedblob.cpp

@@ -239,7 +239,7 @@ TabVector* AlignedBlob::FindVerticalAlignment(AlignedBlobParams align_params,
   box = it.data()->bounding_box();
   int start_x = align_params.right_tab ? box.right() : box.left();
   int start_y = box.bottom();
-  // Acceptable tab vectors must have a mininum number of points,
+  // Acceptable tab vectors must have a minimum number of points,
   // have a minimum acceptable length, and have a minimum gradient.
   // The gradient corresponds to the skew angle.
   // Ragged tabs don't need to satisfy the gradient condition, as they

src/textord/cjkpitch.cpp

@@ -306,7 +306,7 @@ class FPRow {
   // merged with others any more.
   void FinalizeLargeChars();
 
-  // Ouput pitch estimation results to attributes of TD_ROW.
+  // Output pitch estimation results to attributes of TD_ROW.
   void OutputEstimations();
 
   void DebugOutputResult(int row_index);
@@ -528,7 +528,7 @@ void FPRow::OutputEstimations() {
 
   if (good_pitches_.size() < all_pitches_.size() * kFixedPitchThreshold) {
     // If more than half of the characters of a line don't fit to the
-    // fixed pitch model, consider the line to be propotional. 50%
+    // fixed pitch model, consider the line to be proportional. 50%
     // seems to be a good threshold in practice as well.
     // Anyway we store estimated values (fixed_pitch, kern_size, etc.) in
     // real_row_ as a partial estimation result and try to use them in the

src/textord/cjkpitch.h

@@ -57,7 +57,7 @@
 // the final decision about fixed-pitchness of each row and compute
 // pitch and spacing parameters.
 //
-// (If a row is considered to be propotional, pitch_decision for the
+// (If a row is considered to be proportional, pitch_decision for the
 // row is set to PITCH_CORR_PROP and the later phase
 // (i.e. Textord::to_spacing()) should determine its spacing
 // parameters)

src/textord/colpartition.cpp

@@ -647,7 +647,7 @@ ColPartition* ColPartition::SingletonPartner(bool upper) {
 // Merge with the other partition and delete it.
 void ColPartition::Absorb(ColPartition* other, WidthCallback* cb) {
   // The result has to either own all of the blobs or none of them.
-  // Verify the flag is consisent.
+  // Verify the flag is consistent.
   ASSERT_HOST(owns_blobs() == other->owns_blobs());
   // TODO(nbeato): check owns_blobs better. Right now owns_blobs
   // should always be true when this is called. So there is no issues.

src/textord/colpartitionset.h

@@ -101,7 +101,7 @@ class ColPartitionSet : public ELIST_LINK {
   // by the given coords(left,right,y), with the given margins.
   // Also return the first and last column index touched by the coords and
   // the leftmost spanned column.
-  // Column indices are 2n + 1 for real colums (0 based) and even values
+  // Column indices are 2n + 1 for real columns (0 based) and even values
   // represent the gaps in between columns, with 0 being left of the leftmost.
   // resolution refers to the ppi resolution of the image. It may be 0 if only
   // the first_col and last_col are required.