Method and apparatus for proofreading in a computer system5576955Abstract The present invention is a method and apparatus for processing errors in a computer system. The present invention scans information, generates an error unit for each error detected, provides an intuitive interface for communication of errors, and facilitates correction of errors. A data processing computer system of the present invention provides a proofreading capability for addressing errors such as spelling, grammar, usage, punctuation, broken words, doubled words, and capitalization. An error unit is created for each error detected during the proofreading scan of a document. An error unit uniquely identifies each error detected. Further, an error unit retains other information about an error including the error type, location, length, and correction status. The error unit is used to create a list of errors, or Error List. The Error List provides the ability to view errors in one location, and facilitates the error correction and proofreading process. Errors can be addressed in any order. When an error is selected from the Error List, the tight integration between the error and the scanned document provides the ability to locate and highlight the error within the scanned document. A Proofreading Screen provides additional information and proofreading functions. The error selections, and/or all occurrences of an error selection, can be corrected, ignored, learned, or autocorrected. The present invention provides the ability to group errors. Thus, a more intuitive interface is provided to facilitate the proofreading exercise. Further, operator interaction is minimized and simplified. Claims We claim: Description BACKGROUND OF THE INVENTION
______________________________________
error location thirty-two bits
error length sixteen bits
error identifier thirty-two bits
correction status sixteen bits
original error pointer
undo stamp
______________________________________
The error location is an offset from beginning of the proofread text. The correction status indicates whether or not the error has been corrected, learned, ignored, auto-corrected, disabled, or corrupted. The undo stamp is used in when an undo operation is performed. The error identifier consists of multiple fields including a unique error identifier and an error type. The types of errors detected are dependent on the proofreading engine used to scan the text. Possible error types are: spelling, usage, custom usage, punctuation, broken words, doubled words, capitalization, and spacing. The present invention has the capability to address any type of error. The present invention identifies errors as mechanical or non-mechanical errors. A mechanical error is an error that is context-sensitive, and is best understood by viewing the error in the text in which it occurred. Examples of mechanical errors are punctuation, broken words, doubled words, capitalization, and spacing. Non-mechanical errors can usually be communicated without viewing the error in its context. Examples of non-mechanical errors are spelling and usage. Error List Once the text has been scanned for errors, the present invention builds a list of the detected errors at 150 in FIG. 2. The Error List is used to communicate errors (either mechanical or non-mechanical errors) to the user in a more intuitive manner than prior art systems. The Error List provides the ability to communicate the errors outside of the context of the original text. This results in a greater number of errors being displayed on the screen, and provides the ability to scan the errors to determine the manner in which to approach the correction of the errors. Further, errors can be implicitly ignored. That is, an Ignore operation is not needed to skip an error not requiring any action. An Error can be ignored by not selecting the error in the Error List. With the Error List, it is no longer necessary to correct errors in the order in which they are encountered. Errors can be corrected in any desired order. Further, the Error List provides the ability to select errors to be grouped, and to perform a proofreading function on the grouped errors. FIG. 5 illustrates an Error List with suggested alternatives. A user may wish to address a subset of errors first and then address the remaining errors individually or as a group. For example, the user could quickly scan the Error List selecting errors with acceptable correction suggestions. The user could request the correction of these errors as a group. Once the errors are corrected, they are eliminated from the Error List. This process provides the ability to correct errors by skipping over errors and choosing errors in an order other than the order that they were encountered or where they appear in the text. The Error List can be displayed with or without an alternative correction suggestion. While FIG. 5 illustrates an Error List with suggested alternatives, FIG. 8 illustrates an Error List without suggested alternatives. As such, the present invention provides the ability to toggle between an Error List containing suggestion alternatives and an Error List without suggestion alternatives. The box labeled ErrorList Suggestions in FIG. 8 becomes Hide Suggestions when suggestions are included in the Error List as shown in FIG. 5. Mechanical errors are designated by a descriptive phrase to indicate the type of error enclosed in brackets. The "{Spacing}," "{Doubled Word}," and "{Broken Word}" are examples of mechanical error entries in the Error List. Proofreading Exercise Referring to FIG. 2, "Perform proofreading functions on scanned Document" 400 communicates the errors and provides proofreading functions to correct the errors. In the preferred embodiment, error communication and correction are facilitated by the use of three windows or display boxes: a Text Window containing the scanned text document, a proofreading screen containing proofreading functions, and an Error List window. FIG. 5 provides an example of these three windows. The Text Window contains a copy of the text being proofread. The current error is located in the text and highlighted. The current error is either the first potential error detected during the scanning process, or the most-recently selected error. Referring to FIG. 5, "all agree" is the current error, and is highlighted in Text Window with a shaded box. The present invention may be implemented using any form of highlighting technique. Other highlighting techniques includes boldface, italics, reverse-video. The Proofreading Screen includes informational boxes and the proofreading functions. FIG. 5 illustrates a Proofreading Screen in the preferred embodiment. The Message Window 500 provides a descriptive message to further explain the nature of the current error. Suggestions 502 are provided in the box below the Message Window. CorrectWith text 504 indicates the most-likely correction or the last suggestion entered by the user. The Proofreading Screen further provides the following proofreading functions: Correct, Ignore, Learn, Undo, AutoCorrect, Disable, Done, ErrorList, and ErrorList Suggestions/Hide Suggestions. The Correct function corrects the current error in the current context only. Correct All corrects all occurrences of the current error in the scanned text. Ignore discards all occurrences of the current error and the current error is not included as a potential error for the current session (i.e., the current editing session). One occurrence of the current error can be ignored by selecting the next error. Learn adds the current word to the proofreading dictionary. Once a word or phrase is learned, the present invention does not identify it as an error in any subsequent proofreading exercises. Undo cancels the last action. AutoCorrect automatically replaces a word or phrase with another word or phrase. The replacement word or phrase is either the suggestion alternative provided by the proofreading process, or by word or phrase entered in the CorrectWith field. Disable provides the ability to disable an error category such as Usage and Spelling. The Disable command may be used to disable an error type at scan time, or when errors are being viewed in the Error List. The current error's error type is designated as the type of error to be disabled when the Disable button is selected. Disabling an error type prior to scanning a document excludes any error checking for that type of error when the text is scanned. The Disable command can further be used to eliminate scanned error types from the Error List. The current error's error type is designated as the type of error to be excluded from the Error List. Selecting the Undo command after a Disable command is performed on the Error List causes the disabled error type to be reinstated in the Error List. Done ends the proofreading session. The ErrorList button opens the Error List window. ErrorList Suggestions provides the ability to toggle back between an Error List containing suggestion alternatives, and an Error List without suggestion alternatives. Errors can be corrected conventionally by sequentially perusing the scanned text and using the Proofreading Screen. Further, the present invention provides the Error List as an intuitive interface for error correction. Errors can be viewed as a list, and can be corrected in any order. Errors can be grouped, and proofreading functions can be applied to all of the grouped errors. Therefore, errors can be corrected in any desired order. Further, errors can be handled collectively thereby reducing the amount of time it takes for error correction. The Error List is opened when the ErrorList button is chosen from the Proofreading Screen (i.e., ErrorList in FIG. 5). When an error is selected in the Error List, the error is located in the scanned text. The text window scrolls to error within the scanned text. FIG. 5 illustrates the condition of Text Window, Proofreading Screen, and Error List at the inception of the proofreading exercise. The current error in the Error List (i.e., "all agree") is highlighted with a shaded box. The current error is also highlighted in Text Window. Thus, the user may view the error in the conventional manner used in prior art systems and in the Error List in the present invention. In Proofreading Screen, Message Window 500 contains a description of the current error and Suggestions 502 provides a correction suggestion. The CorrectWith text 504 indicates the replacement text that will replace the "all agree" error, when the Correct or Correct All function is invoked. FIG. 6A illustrates the condition of the screen is the Correct function is selected to correct "all agree" with "agree." In Text Window, the phrase "all agree" is replaced with "agree." "All agree" is removed from the Error List. The next error in the Error List (i.e., "{Spacing}") is highlighted in the Error List. The spacing error associated with the Error List entry is located in the text in Text Window, and the associated text (i.e., "mistakes.") is highlighted in Text Window. Proofreading Screen is updated to reflect the most-recent current error. Message Window 500 contains descriptive information about the current error, Suggestions 502 contains correction suggestions, and CorrectWith text 504 contains the most-likely suggestion alternative (i.e., "mistakes."). The present invention provides the ability to group errors and to perform operations on a group of errors such as correcting all occurrences of each of the errors in the group; excluding all occurrences of each of the errors in the group as potential errors during the current session; adding each error to the dictionary. Groups of errors may be identified using the Error List. Contiguous errors and non-contiguous errors can be grouped together. The Correct, Ignore, and Learn functions may be used to operate on a group of errors selected from the Error List. A correction can be made to all of the errors included in a group. Further, group of errors can be bypassed by selecting Ignore after identifying a group of errors from the Error List and selecting a range of text in the scanned text. A group of errors can be added to the dictionary after identifying a group of errors. Errors may also be selected or grouped by selecting a portion of text in the Text Window. The desired portion of the scanned text is highlighted in the text window. The errors in the highlighted section of the text will be highlighted in the Error List. The Correct, Ignore, and Learn proofreading functions can be used on any or all of the highlighted errors. The present invention further provides the ability to address the errors in any order. The errors in the Error List can be addressed in any order, and the Error List can be perused multiple times to including additional errors in a group before performing an operation. FIG. 6B illustrates the condition of the screen, if the spacing error is bypassed to address a group of errors with acceptable correction suggestions. The initial perusal of the Error List might yield a group of errors including "Never the less," "amny," "youll," "suchan," and "there-after." A check mark is placed to the left of each group member after selection. Although a check mark is used in the preferred embodiment, implementation of the present invention may successfully use any means of highlighting group members. Additional scans might result in further selecting "In the majority of cases," "be sure and," and "readed." FIG. 6C illustrates the Error List after subsequent passes of the Error List. A check mark is placed to the left of the additional group members' entries. Once all of the group members have been selected, a proofread function can be selected to address the errors in the group. FIG. 7A illustrates the condition of the screen after the Correct All function is selected and performed to address the errors in the group. All occurrences of each group member is corrected in the text and the display in Text Window. Each occurrence of a group member has the same error identifier. Thus, each occurrence of a group member can be found, and each occurrence location in the text located and modified. Further, all occurrences of each group member is eliminated from the Error List. Message Window 500, Suggestions 502, and CorrectWith 504 in Proofread Screen is updated to reflect a new current error (i.e., "{Spacing}" in the Error List and "mistakes ." in Text Window). Since the present invention provides the ability to address errors in any order, the spacing error can be bypassed to select any other error in the Error List as the current error. For example, the "{Broken Word}" error can be selected instead of addressing the spacing error. Once the broken word error is selected, it becomes the current error, and the display is updated to reflect the most-recent current word. FIG. 7B illustrates the condition of the display after selecting the broken word error. The current error is located in the text and scrolled to and highlighted in Text Window. This provides the opportunity to view the current error, a mechanical error, in its context within the text. Proofreading Screen is updated with an associated message in Message Window 500, a correction suggestion for the current error in Suggestions 502, and the most-likely correction suggestion in the CorrectWith text 504. The "{Broken Word}" entry in the Error List is highlighted. The current error may be addressed, another error can be selected as the current error, or a group of errors can be selected to be manipulated by the proofreading functions. Building ErrorList Process Flow Referring to FIG. 2, once the text has been scanned and error units have been generated for the error detected during the scanning, the Error List is built. FIG. 3 illustrates the BuildErrorList process flow. At decision block 252 (i.e., "all error units processed?"), if all of the error units generated while scanning the text have been processed, processing ends. If not, the next error unit is retrieved at processing block 254. At processing block 256, the error type is filtered from the error identifier. At decision block 258 (i.e., "mechanical error?"), the error type is examined to determine if the error is a mechanical error (i.e., a context-sensitive error). If it is a mechanical error, the phrase used to designate the error type is stored in the next position in the Error List at processing block 260, and processing continues at decision block 252 (i.e., "all error units processed?"). If the error is not a mechanical error, the error word of phrase is copied from the scanned text to the Error List at processing block 262. At processing block 264, a list of correction suggestions is generated and the first entry in the list is identified as the most-likely correction. At decision block 266 (i.e., "suggestions displayed?"), if the suggestions' toggle switch is set to hide correction suggestions, processing continues at decision block 252 (i.e., "all error units processed?"). If the suggestions' toggle switch is set to display correction suggestions, the error word or phrase in the Error List is appended with an arrow and the most-likely correction (i.e., the first correction suggestion in the list of correction suggestions for the error) at processing block 268. Processing continues at decision block 252 (i.e., "all error units processed?"). Proofreading Exercise Process Flow Referring to FIG. 2, after the Error List has been generated, the Text Window and Proofreading Screen is displayed. If ErrorList is selected, the Error List is displayed. The proofreading functions are selectable from the Proofreading Screen. FIG. 4A illustrates the proofreading process flow. At processing block 402, the current error is located and highlighted in the text in Text Window. The current error is the first error encountered in the scanning process, or the most-recently selected error. The Message Window is filled with a descriptive message regarding the current error, and the Suggestions Window is filled with the most-likely correction at processing block 404. The most-likely correction is the first correction suggestion in the list of suggestion alternatives previously generated. CorrectWith is filled with the most-likely correction at processing block 406. If the Error List is not opened at decision block 408 (i.e., "Error List opened?"), processing continues at decision block 412 (i.e., "proofreading function selection?"). If the Error List is opened, the current error is located in the Error List and highlighted at processing block 410. User input may be in the form of a proofread function selection, error selection, or correction word or phrase. If the input is not a proof read function selection at decision block 412 (i.e., "proofreading function selected?"), CheckOtherInput is invoked at processing block 420. If a determination is made that a new error has been selected as the current error at decision block 444 (i.e., "current error selection?"), processing continues at processing block 402. Otherwise, processing continues at decision block 412 (i.e., "proofreading function selected?"). If the input is a proofreading function at decision block 412 (i.e., "proofreading function selected?"), DoProofreadingFunction is invoked at block 450, and processing continues at decision block 412 (i.e., "proofreading function selection?"). FIG. 4B illustrates the flow of DoProofreadingFunction invoked in the PerformProofreadingFunction flow. If an Undo operation is identified at decision block 452 (i.e., "Undo?"), the last proofreading operation is undone at processing block 454 and processing continues at return block 494. If an Undo operation is not identified at decision block 452, processing continues at decision block 456 (i.e., "AutoCorrect?"). If an AutoCorrect operation is identified at decision block 456, the error is replaced with the CorrectWith value at processing block 458. Further, the auto correction value is stored in the scan dictionary at processing block 458. At processing block 464, all occurrences of the selected error(s) are corrected, and processing continues at return block 494. If an AutoCorrect operation is not identified at decision block 456, processing continues at decision block 460 ("Learn, Ignore, Correct, or CorrectAll?"). If a Learn, Ignore, Correct, or CorrectAll operation is identified at decision block 460, a determination is made whether the operation is a CorrectAll operation at decision block 462 (i.e., "correct all?"). If it is not a CorrectAll operation, the selected proofreading function operation is performed on a selected error or group of errors at processing block 466. If it is a CorrectAll operation, all occurrences of the selected error or errors are corrected at processing block 464. In both cases, processing continues at return block 494. If a Learn, Ignore, Correct, or CorrectAll operation is not identified at decision block 460, processing continues at decision block 468 ("Disable?"). If a Disable operation is identified at decision block 468, the error type indicated by the Disable button (i.e., the current error type) is disregarded at processing block 470. A Disable operation will result in the error type being excluded from the scan process, and excluded from the Error List. After excluding the current error type, processing continues at return block 494. If a Disable operation is not identified at decision block 468, processing continues at decision block 472 ("ErrorList?"). If an ErrorList operation is identified at decision block 472, a determination is made whether the Error List is displayed at decision block 474 (i.e., "ErrorList displayed?"). If it is not displayed, the Error List is displayed at processing block 478. If it is displayed, the Error List is closed (i.e., removed from the display) at processing block 476. In both cases, processing continues at return block 494. If an ErrorList operation is not identified at decision block 460, processing continues at decision block 480 ("ErrorList Suggestions?"). If an ErrorListSuggestions operation is identified at decision block 480, a determination is made whether suggestions are displayed in the Error List at decision block 482 (i.e., "ErrorList Suggestions?"). If suggestions are not displayed, suggestions are added to the Error List at processing block 484. If suggestions are displayed, suggestions are removed from the Error List at processing block 486. In both cases, processing continues at return block 494. If an ErrorListSuggestions operation is not identified at decision block 480, processing continues at decision block 488 ("Done?"). If a Done operation is identified at decision block 488, the proofreading session is terminated at block 490. If a Done operation is not identified at decision block 488, invalid input may have been received. Communication of and recovery from this error state is performed at processing block 492. Processing continues at return block 494. At return block 494, processing returns to decision block 412 (i.e., "proofreading function selected?") in PerformProofreadingFunction. As previously described, if a proofreading function is not identified at decision block 412 in PerformProofreadingFunction, CheckOtherInput is invoked at processing block 420 of PerformProofreadingFunction. FIG. 4C illustrates the process flow of CheckOtherInput. Errors may be selected in the Text Window (i.e., by highlighting a portion of the text), or in the Error List. If the input is not an error selection at decision block 422 (i.e., "error selection?"), an inquiry is made at decision block 424 (i.e., "CorrectWith entry?") to determine whether the input is a CorrectWith field entry. If it is not, invalid input may have been received. Communication of and recovery from this error state is performed at processing block 428, and processing continues at return block 442. If CorrectWith input is identified at decision block 424, the CorrectWith field is updated at processing block 426, and processing continues at return block 442. If an error selection is identified at decision block 422 (i.e., "error selection?"), processing continues at decision block 430 (i.e., "part of a group?"). If the error selection is not part of a group of errors, the current error is set to the error selection at processing block 432, and processing continues at return block 442. If the error selection is identified as part of a group at decision block 430, the error type is filtered from the error identifier at processing block 434. At decision block 436 (i.e., "mechanical error?"), if the error is a context-sensitive error, an error is generated as shown in block 438, and processing continues at return block 442. If the error is not a context-sensitive error, the error is marked as part of the group at processing block 440, and processing continues at return block 442. At return block 442, processing returns to decision block 412 (i.e., "proofreading function selected?") in PerformProofreadingFunction. The present invention provides an intuitive interface for correcting errors generated during a proofreading exercise. Errors are enumerated in the Error List thereby allowing error to be addressed in any order, or in groups. Errors can be excluded from the Error List, and errors within a subset of the text can be highlighted for easy identification and handling.
|
Same subclass Same class Consider this |
||||||||||