Text processing apparatus with editing of stored document at each keystroke

Abstract

A text terminal in which a text stream received on a communication link is segmented by a segmenter into chained variable-length segments which are stored in a random access text store. Each segment can contain at most one extended formatting command. Segments have slack areas to allow the subsequent insertion of text. A text editor receives keystroke signals from a keyboard and modifies the stored text stream, new segments being created as necessary for extended formatting commands or when slack areas are full. An interpreter/formatter is invoked when a received text stream has been segmented or when the stored text stream has been modified to format and map the stored text stream into a refresh buffer driving a cathode ray tube (CRT) display. The interpreter/formatter maintains a status index with condensed versions of the extended formatting commands stored in the text stream. Formatting is inhibited by the editor if it receives a keystroke requiring modification of the stored text. After editing segmented text is desegmented by a desegmenter and returned to the host along a communication link in the form in which it was received. Preferably, the terminal is implemented with a microprocessor which can also control standard data entry applications as well as text processing.

Claims

Having thus described our invention, what we claim as new and desire to secure by Letters Patent is:

1. A text processing apparatus for use with a host processor capable of receiving, storing, and transmitting previously keyed text, the text including symbolic characters and text control characters, the text control characters including multi-character extended formatting commands affecting the format of a document, comprising

a keyboard for entering text,

means for receiving text from the host processor in text stream form,

means coupled to said text stream receiving means for storing said text stream, including said control characters, in said data stream format, editing means connected for receiving keystroke signals from said keyboard and including means for modifying said text stream stored in said storing means in accordance with said keystroke signals on a keystroke-by-keystroke basis, and

a display for displaying said stored text,

wherein the improvement comprises,

text segmenting means for dividing said received text stream into segments and for chaining said segments,

each segment having a variable-length of text and having a predetermined maximum length of text, the area within the maximum length not occupied by text being designated a slack area,

each segment having at most one extended formatting command therein and segments not having an extended formatting command therein having a slack area not containing data and therefor available for inserts, and

formatting means for reading said segmented text stream from said text store and mapping the contents thereof onto said display and means in said editing means for invoking said formatting means when a text editing step has been completed.

2. Apparatus as claimed in claim 1, wherein said formatting means includes means to maintain a status index containing extended formatting commands stored within said text store.

3. Apparatus as defined in claim 2, wherein said status index includes a page index table for each page of text stored in said text store and a row index table for each row of the page of text being displayed.

4. Apparatus as defined in claim 1, wherein said editing means includes means to interrupt said formatting means upon receipt of a keystroke from said keyboard invoking said editing means to perform a text editing operation.

Description

BRIEF SUMMARY OF THE INVENTION

An object of the present invention is to provide a text processing apparatus in which text is stored in variable-length storage segments as a text data stream with forward and reverse chaining of the storage segments. Another object is to provide text processing apparatus in which editing, formatting and storage management functions may be performed by a single microprocessor without undue undesirable impact on the performance thereof.

According to the present invention, a text processing apparatus comprises (1) a keyboard for entering text, (2) means for receiving text consisting of symbolic and control characters including extended formatting commands in text stream format, (3) means for storing said text stream including said control characters in said data stream format, (4) editing means for receiving keystroke signals from said keyboard and modifying said stored text stream in accordance therewith on a keystroke-by-keystroke basis, and (5) a display for displaying said stored text in accordance with text control characters contained therein, characterized in that said apparatus further includes (6) text segmenting means for dividing said received text stream into chained segments of variable-length having a predetermined maximum length with each segment having at most one extended formatting command therein and segments not having an extended formatting command therein having a slack area, and (7) formatting means for reading said segmented text stream from said text store and mapping the contents thereof onto said display.

The foregoing and other objects, features and advantages of the invention will be apparent from the following more particular description of a preferred embodiment of the invention is illustrated in the accompanying drawings.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a block diagram showing the component parts of a text processing apparatus according to the present invention;

FIG. 2 shows the structure of a text storage segment;

FIG. 3 diagrammatically illustrates how a text data stream is processed by the apparatus of FIG. 1;

FIG. 4 shows a modification of the apparatus shown in FIG. 1 which can be used for processing non-textual data as well as text; and

FIG. 5 is a block diagram of a preferred embodiment of the invention employing a single microprocessor to edit text stored as a text data stream.

DETAILED DESCRIPTION

An Overview - FIG. 1

In FIG. 1, a text processing terminal comprises a keyboard (KB) 1 by means of which an operator can enter, delete or change text including symbolic characters and formatting characters in a text data stream stored in a text store 2 constituted by a random access memory. Text stored in the text store 2, or a part of the stored text, is displayed on a display screen 3, conveniently constituted by a cathode ray tube (CRT) display. A document to be displayed and/or edited on the terminal is received on communication line 4 from a host processor, not shown, by a segmenter (SEG) 5 which, in conjunction with a store manager (SM) 7 divides the text stream into variable length segments stored in store 2. To display the text on the display 3, an interpreter/formatter 6 maps the text stream into a refresh buffer (RB) 8 from which the CRT display can be periodically refreshed.

A keystroke processor (K/SP) 9 receives keystrokes from the keyboard 1 and invokes a text editor (TE) 10. The editor 10 causes modification of the text data stream stored in store 2 and the changed text will subsequently be formatted and mapped into the refresh buffer 8 for display on the screen 3. During the segmenting and formatting processes, the interpreter/formatter 6 maintains a status index 13 in random access memory. The status index 13 consists of page indices, one for each page of text in the store 3, a row index for each line of text of the current page, and a current index for the current line of text displayed on the screen.

The segmenter 5 maintains a segment number counter, not shown, to identify newly created segments within the store 2. This counter is also available when new segments are created during a subsequent text insertion.

Processing by the interpreter/formatter 6 is initiated either by the segmenter 5 on completion of transferring incoming text to the store 2 or by the text editor 10 on completion of an update to the text in the store 2.

The text editor 10 employs a two-step process. Firstly, it invokes the interpreter/formatter 6 to determine the physical cursor position relative to the text in the store 2. At the end of this phase, a corresponding logical cursor position is known within the store. Secondly, the particular operation on the text is performed (insertion/deletion) at the logical cursor position and the interpreter/formatter 6 is re-invoked to display the effect of the change on the screen.

The store manager 7 uses the VMT storage primitives unchanged. All stored text is referred to using the VMT bipartite addressing (segment identifier plus offset).

The Text Segment Header - FIG. 2

As shown in FIG. 2, a text segment 14 includes a 6-byte header field (HDR) 15 which contains the segment identification and its length, a 2-byte locator field (LOC) 16 which identifies a storage location containing the real address of the segment, a 2-byte field (PRV) 17 which identifies the previous segment, a 2-byte field (NXT) 18 which identifies the next segment, a 1-byte field (L) 19 which contains an indication of the length of the text stream stored in the segment, a variable-length field 20 containing the text data stream stored in that segment, and a field 21 forming a slack area of spare space into which the text data field 20 can expand as a result of insertion. The header field 15 and locator field 16 are used by the store manager 7, FIG. 1, to fetch and store segments within the store 2. This storage management process is described fully in the aforementioned European Patent application No. 80 3021922, entitled, "Virtual Memory Terminal" and will only be briefly referred to in this specification. The fields 17 and 18 are used by the interpreter/formatter 6, FIG. 1, to obtain the previous and succeeding portions of text from the store 2 (using the store manager 7). The text length field 19 enables the text editor 10 to determine quickly whether the segment is empty, that it has no text within it, or full, that is has no slack space within it.

The Text Data Stream - FIG. 3

Before describing the segmenting, editing and formatting processes in more detail, reference will be made to FIG. 3 which pictorially illustrates how a text data stream 22 is segmented, edited and formatted to appear on the display screen. It will be appreciated that in practice the text stream will consist of many hundred of characters. It is preferred that the text stream is in the form of the IBM SCS text stream although this is not essential to the present invention. The text stream 22 consists of a number of text characters consisting of text symbols and imbedded text commands. As indicated above, the text commands consist of two types, multiple-character commands, within this specification called extended formatting commands, and single-character commands. Examples of different types are given above and in the aforementioned Reference Manual. The text stream 22 shown in FIG. 3 consists of symbolic text in the form of alphabetic characters and text commands. Of the three text commands shown, those represented by the symbols "<" and ">" represent the single-character commands "new-line" and "required-new-line" respectively and that represented by the symbol "!" represents the extended formatting command "temporary-left-margin".

As illustrated in FIG. 3, the segmenter 5 (FIG. 1), in conjunction with the store manager 7, divides the text stream into three segments stored within the store 2, one consisting of a stream of text including one single-character text command, one consisting of an extended formatting command and one consisting of text and three single-character text commands. Each segment has header information and a slack area. The rules of segmentation are such that each segment has a maximum length of 256 bytes which means field 19, FIG. 2, need only be one byte. (This is significant in byte organized machines). It will be appreciated, however, that the maximum length is optional and can be made larger or smaller as desired for the particular storage system used. Any extended formatting command, recognized because its first byte is a Control Sequence Prefix (CSP), always starts a segment. The only other character which can be stored in the same segment is a trailing blank or new line character immediately following the extended formatting character. Thus, a segment will contain either none or only one of the extended formatting commands.

The Extended Formatting Command

As is explained in the above manual, each extended formatting command consists of the CSP, followed by a class byte, a count byte, a type byte, and, optionally, parameters. The class and type bytes identify the command, the count byte contains the length of the command, starting with and including the count byte itself.

The interpreter/formatter 6 loads a condensation of extended formatting commands into the status index 13 to enable efficient and rapid mapping of the stored text stream into the refresh buffer for display on the screen. Using the store manager 7 and status index 13, the interpreter/formatter 6 reads the text stream from the store 2 and maps the symbolic text and commands onto the display screen 3 formatted in accordance with the document formatting commands contained within the text stream (and in the index 13). Two display modes are possible, one without the symbols representing the text commands, and, as represented by 3', with the text commands displayed. This latter mode is particularly useful when an operator is editing the document since it shows how the document was originally created. During the editing process, the editor 10 causes characters to be deleted or inserted from or into the text stream stored within the store 2. As will be seen later, the editing process may result in the need to create new segments and to remove old segments.

What has been described so far is a text processing terminal which is capable of receiving a text data stream from a remote processor, segmenting and storing the received text stream within a store, formatting and mapping the stored data stream (or a part thereof) on a display and editing the stored data stream. The text data steam received from the host may be a complete document consisting of one or more pages, or it may consist of part of a document. Where only part of a document is received, any extended formatting commands which affect that part of the document will also be received. The received extended formatting characters are stored as part of the text data stream but are also loaded into the status index to control the formatting on the screen of that part of the received document which they govern or control.

The interpreter/formatter 6 initializes the screen by building a table of addresses of each screen row. This table is used to speed the search for the physical cursor during editing. The number of rows which can be displayed is determined by the particular display used. If an IBM 8775 Display Station or similar display is used, parameters can be supplied by the host to configure the "size" of the screen--this not the physical size, but the number of rows and columns which can be displayed. The screen is cleared with the cursor positioned at the start of the document display area.

Status Index 13

The interpreter/formatter 6 maintains the status index 13 to ensure rapid formatting. As indicated above, the status index 13 includes a page index which allows rapid access to a particular page without the need to scan the preceding text data stream, a row index which contains similar information about each row of the current page, and a current index which is used by the interpreter/formatter to collect the text formatting commands prior to transferring them to the appropriate row index.

The appearance of the page index is as follows:

1. Page width

2. Page depth

3. Temporary Left Margin position

4. Tab settings

5. Address of data within the store (segment+offset)

6. Flags (skip, adjust, center)

7. Page number

Optionally, the index could hold the address of the tab setting rather than hold the tab settings themselves. This could save both process time and memory usage.

The row index contains one entry for each text line of the current page. The entries are similar to those in the page index except that the page number is not required. The first row index entry contains status information for the first line of a page. Alternatively, the row index entries could be limited to a single screen of text lines. This would use less storage but lead to poorer browsing performance since to scroll back one screen would require the top line row index entry of the current screen to be saved. (The interpreter/formatter would commence processing at the beginning of the first line of the current page building a row index entry for each line until the entire first screen is displayed, a comparison then being made between the bottom entry of the resulting row index and the original stored entry. If these match the "screen back" operation is complete, but if not, then the whole process would be repeated commencing with the next line of the page and so on until the test is successful. This complication arises because the displayed lines may not necessarily be on an integral screen boundary within the page. This will almost certainly be the case if the operator has used a cursor scroll key.)

The current index is a single entry containing the same information as the page index. It is used for holding the current document status information derived during the formatting operation. The index for the first page may be initialized with a predetermined set of data (tab settings, temporary left margin setting, etc.) or alternatively, the initialization data could be sent to the terminal from the host processor.

A Modified Apparatus - FIG. 4

FIG. 4 is a modification of the terminal shown in FIG. 1 showing how the text processing apparatus of the present invention could be implemented on a data-entry display terminal such as the IBM 8775 Display Terminal. In FIG. 4, outbound data received from the host processor is processed within an outbound data stream processor (ODSP) 23 which routes the incoming data processing functions (DP) 24. Routing is determined by the type of header associated with the data. When in data processing mode as opposed to text processing mode, keystroke processor 9 routes keystrokes to the normal DP functions 24. Text or data streams being sent back to the host are routed through an inbound data stream processor (IDSP) 25 as shown. Those parts of FIG. 4 which are used only for text processing are shown in phantom in FIG. 4 while parts which are used for data or data and text are shown in solid line.

The Embodiment of FIG. 5

FIG. 5 illustrates how the apparatus of FIGS. 1 and 4 can be built using a single microprocessor 26, for example, a Motorola M6800 microprocessor, with various components, known per se, connected thereto through an 8-bit data bus 27, a 16-bit address bus 28 and an interrupt line 29. The keyboard 1 is connected to the system bus through a keyboard adapter 30, which, in known manner, interfaces the scan codes from the keyboard with the microprocessor 1. A communication adapter 31 interfaces the microprocessor 26 with a communication link 32 to the host processor. Connected to the data and address buses 27 and 28 is a random access memory 33 divided into a number of sections. Section 34 contains microcode for data entry applications. Section 35 contains data entered in the terminal or sent by the host. Section 36 constitutes the text store 2 of FIG. 1 in which the text is stored in text stream form. Section 37 contains code for managing text segments in the text store 7, FIG. 1, and section 38 contains code for controlling the segmenting, desegmenting, formatting and editing processes 5, 11, 6 and 10, FIG. 1. Section 39 stores the status index 13, FIG. 1 containing the condensed extended formatting commands. Finally, section 8 constitutes the refresh buffer for refreshing the CRT display 3 as in FIG. 1. Data or text to be displayed is loaded into section 8 from section 35 or section 36 by the microprocessor when the data or text is called by the operator or a program. Refresh logic 40 periodically reads the contents of section 8 and drives the CRT display 3 in accordance with the characters stored therein. A read only store (ROS) 41, is used to store common code which does not require changing.

As indicated above, the various components of FIG. 5 are known per se and may be constituted by suitably configuring an IBM 8775 Display Terminal. No detailed description of the components will be given. However, it is in order for more details to be given of the priorities assigned to the various processes and for flowcharts of the more complicated processes to be given.

The task structure is similar to that on the 8775 and has six levels of priority as follows, highest priority first:

    ______________________________________
    1.      Host link communication.
    2.      Keyboard adapter providing keyboard data
            service
    3.      Keystroke processing, text editing,
            interpreting and store management.
    4.      Inbound (to the host) data stream processing,
            desegmentation.
    5.      Outbound (from the host) data stream
            processing, segmenting.
    6.      Monitoring, scheduling.
    ______________________________________

                  TABLE I
    ______________________________________
    (SEGMENTATION PROCESS)
    ______________________________________
    1.    Interpret first character in text data stream,
          go to 2.
    2.    If extended formatting command, go to 3; if not,
          go to 8.
    3.    Create new segment, store command, and go to 4.
    4.    Interpret next character in data stream and if new
          line command store with extended formatting command
          and go to 7; if no, go to 5 if it is not an extended
          formatting command, but to 2, if it is.
    5.    Create new segment and store character, go to 6.
    6.    If another character in text data stream go to 7,
          if not go to 11.
    7.    Interpret character and go to 2.
    8.    If existing segment at maximum go to 9; if not,
          go to 10.
    9.    Create new segment, store character and go to 6.
    10.   Store character in existing segment and go to 6.
    11.   Terminate segmenting process.
    ______________________________________

                  TABLE II
    ______________________________________
    (CHARACTER INSERTION PROCESS)
    ______________________________________
    1.     If character being inserted is extended formatting
           command go to 2; if not, go to 3.
    2.     Create new segment, and modify pointers.
    3.     If insertion point is first character position in
           segment, go to 4; if not, go to 5.
    4.     If slack available in previous segment, fetch
           and store inserted character; if not, go to 5.
    5.     If slack available in current segment, insert
           character; if not, go to 6.
    6.     Request new segment and if sufficient space in
           store available, go to 10; if not, go to 7.
    7.     Consolidate text segments in storage and go to 8.
    8.     If sufficient space available, go to 10; if not,
           go to 9.
    9.     Indicate storage full or use backing store.
    10.    Create new segment, insert text in existing
           segment to right of insertion point in new
           segment, delete text to right of insertion point
           in existing segment and insert new character at
           end of existing segment, modifying pointers in
           existing and next segments, and length fields
           in existing and new segments.
    ______________________________________

    ______________________________________
    (a)     Page Boundary
    (b)     Required New Line
    (c)     Extended formatting command and terminal in
            `display controls` mode.
    (d)     New Line and terminal not in `Adjusted Text`
            mode.
    (e)     The right hand margin of the document is
            reached.
    (f)     Skipping is in effect.
    (g)     End of Text Stream encountered.
    (h)     The logical cursor has been found as requested
            by the editor 10.
    (i)     In `no display controls` mode an extended
            formatting command is found while formatting
            to the right of the left hand margin.
    ______________________________________

• Inventors
  Lawrence, Stephen G.; Middleton, Brian H.;
• Assignee
  International Business Machines Corporation (Armonk, NY)
• Published
  Aug-7-1984
• Current US Classes:
  400/63
  715/529
• Application #
  273273
• International Classes
  G06F 009/00
• Field of Search
  364/200 364/900
• Examiner
  Gruber; Felix D.
• Agent
  Robertson; W. S.
• US Patent References:
  3972026
  4040024
  4203107