The Newt's Cape user interface can be quite confusing due to its dual purposes: for creating/viewing books (applications), and browsing web pages. Before delving too deeply into details, you may want to look at two sample scenarios, which highlight typical commands and preferences.
Newt's Cape appears initially as a 2-line floating "control panel" at bottom of the screen (over the book control bar area): a single-line URL and a row of command buttons. Interface elements appear below, more or less in left-to-right and top-to-bottom order. Note: The location/appearance of interface items depends on MessagePad model, screen size/orientation, installed functionality, and book status. Commands that are not relevant to the current context or system configuration are surrounded by parentheses, e.g., (Open Newtworks); if selected, they should do nothing. Tap on this book's overview button to navigate interface items.
move Newt's Cape by dragging the dot at top center. Instead of moving it, you can also hide Newt's Cape temporarily, and redisplay it later from the Newt's Cape icon in book control bar or from Extras.
This triangle button expands the single URL line to 3 lines; toggling it again collapses back to a single line.
Enter a URL in this area. A caret should automatically appear. Drag&drop text, or use handwriting or a keyboard to enter a location. You can also highlight a URL in Notes (or NewtWorks), and select Newt's Cape:getURL from the Notes action menu (Newt's Cape from the NewtWorks Tool menu). You can select items from Bookmarks. During processing, the area displays the current document id (URL or ISBN). Typical values:
Tap keyboard icon to pop up a keyboard. 2.x: Newt's Cape adds default (http://, www., .com, .edu, .net, .htm, .html) or user defined keys to top row of keyboard -- for standard system keyboard, . for easier replace, it does Select All of current text view
If you hit Return key (on soft or hard kbd) in Newt's Cape's URL field, this is same as File:Open Location (slight delay); automatically closes soft keyboard
If you have a hard keyboard attached, there are several keyboard equivalents available:
Tap the Newt's Cape icon in order to open the current URL. This is same as File:Open Location. While Newt's Cape is transferring or processing a document, the icon changes to a Stop icon (up-raised hand) -- tap to stop the current operation. (If numerous operations are queued, e.g., autoloading images, you can tap Stop multiple times, or use File:Reset HTTP).
This button can be customized via i:General:Other Options:Newt's Cape button to open Bookmarks or HTML cache. In the book bar, the same icon redisplays Newt's Cape if hidden.
This triangle button expands/collapses status line, error line and Stop button under the row of command buttons.
display Newt's Cape version and short description in HTML, and automatically convert this into a book. This also appears automatically when you start Newt's Cape if you haven't registered and paid yet.
Open the Newt's Cape Introduction/TOC book, if installed; or another book if a custom helpISBN specified.
If the Newton Register application is installed, opens a registration screen. This is a convenient and reasonably safe way to register with a credit card via email. Newton, Mac, Windows, web versions of Register After you've paid, be sure to turn on the I paid checkbox to disable the automatic About box.
Preferences (General, Appearance, Encoding, Process, etc.) appear directly in the i menu -- these are covered a little later
This button and remaining command buttons are intended to mimic the commands/menus you might find in a typical desktop web browser.
It's probably more convenient to use the Newt's Cape icon button to open the current URL. You can also use the Return key on keyboard.
Newt's Cape scans Notes(Notepad) for text entries that begin with <HTML> -- this may take a few seconds the first time. Newt's Cape pops up a list of TITLEs (or "untitled" if no TITLE tag) of possible documents for you to select. Newt's Cape ignores anything before <HTML> (or <!DOCTYPE) and anything after </HTML>. In the Notes application, you can select Newt's Cape:Open from the action menu.
Note: The popup menus for Notes, Inbox, Paperback and Save as Text to Desktop do not scroll on 1.x (unless you use a third party utility). File:Open Notes, Inbox, Paperback, WabumiUni(~J) may not indicate initially whether items are available; however, after selecting the command (and scanning for items), cached results are used the next time to "disable" menu item (if no items are available) or speedup selection for subsequent picks.
Pops up a menu of Inbox (via email, beam) entries that contain HTML documents (like those described for Notes); if you use EETransfer or newtFTP, you can include HTML documents, GIF and JPEG files, or files handled like helperApps like PilotDOC and ModSaver. If a large HTML document was split into multiple messages, Newt's Cape displays the title from the initial message, and combines the rest (tested with Aloha). Alternatively, you can Put Away items from the Inbox to the Notepad, and then edit/process from there, or you can select Newt's Cape:Open from the Inbox action menu.
Pops up a menu of Paperback "books" that contain HTML source documents. This can be a convenient way to transfer and access large, static HTML documents and not clutter up Notes, heap or cache. In addition to containing an <HTML> tag, the text should not contain any null (0x0) characters -- otherwise, a Paperback may not appear in the popup menu.
2.x, Japanese(~J) version only. WabumiUni is a Paperback-like application especially for Japanese-encoded characters. It should be available on a UniFEP CD-ROM (that comes with the Japanese Newton) from Enfour.
2.1 only. If NewtWorks is available, pops up a menu of documents that contain <HTML>. The NewtWorks titles are listed (these may or may not be the same as the HTML source TITLE). Note: this interface relies on undocumented APIs, so is not guaranteed to work. HTML source is processed like Notes. See NewtWorks HTML via Summary preference.
2.x only, if cache on. display an "overview" dialog that contains checkboxes, URLs and other info for current HTML documents, and filing and routing buttons. the view is less than full screen for NOS 2.1 when NIE is connected to avoid system hangs. Note: the following description (with some exceptions) also applies to File:Open Image Cache, Bookmarks:Browse/Schedule, View:Load with Images, View:Load Frames, View:Load Links
2.x only, if cache on. opens overview of current image (GIF,JPEG) URLs, similar to HTML Cache. Note: occasionally (depending on command/context), if a file doesn't have appropriate file extension, its cached entry might not be found.
2.x only, if NetHopper installed. pops up a menu of URLs of current, unconverted HTML and GIF source files, typically stored in its cache by WebXPress. Note: once you view a cached HTML source with NetHopper, it is converted to a different internal format, so source is no longer available; converted GIFs should still work, though "scale to fit?" may not work correctly (and animated GIFs would not work). Note: you should be able to access a URL (for document stored in NetHopper cache) directly via File:Open Location and via links from other documents.
If you have a cable connected to your Newton, and Sloup 2.2a (or greater) installed, you can access HTML documents via serial or AppleTalk directly from your desktop computer and a terminal program. Select a serial or ADSP option from the popup menu. Once you have opened a connection, this item changes to Close Desktop. You can transfer HTML documents separately via Sloup and access from Notes or Newtworks; or HTML/GIF via EE Transfer and access from Inbox.
Transfer an HTML file via your terminal emulator using a "Send Text" command. For 1.x, Newt's Cape parses each line immediately. For 2.x, Newt's Cape collects all of the lines before parsing, and if you have desktop? cache option selected and a BASE HREF (URL) in document, source is stored in HTML cache; you can also send non-HTML text files if you include header info.
Newt's Cape automatically disconnects after encountering a complete line with "</HTML>" or usually if it detects "Out of Heap". Otherwise, select Close Desktop after the transfer is complete. You might also open a connection to log error messages while processing documents from another source, e.g., Notes or to support the HTMList Save as HTML to Desktop command.
Many terminal programs will work for serial, such as ZTerm on Macintosh or Terminal on Windows. You should configure it as follows:
Note: Newt's Cape can handle Macintosh, DOS and Unix end-of-line characters. To avoid overwhelming Newt's Cape and losing text, make sure that your terminal emulator sends one line at a time and supports XON/XOFF; you may also wish to turn off line wrapping. For further details about configurations, see the the usage scenario and Sloup documentation.
Once you have serial working at 9600, you may wish to try 57600 -- this will be somewhat faster, but HTML processing (on 1.x) will likely be a bottleneck, especially if Display Item Process option is selected.
For ADSP (AppleTalk protocol), your desktop system will need an application such as ClarisTerm that supports ADSP ("CommToolBox-aware"). Select "Wait For Connection" in your terminal program, then select Newt's Cape's ADSP option -- you will then be offered a chooser to select a network location. Last ADSP zone and host saved as preference; confirms as default for next connect. Note: ADSP may not be too reliable on older Newtons.
The connection is normally closed automatically when </HTML> is encountered (there must be a final line-end character also). Select this (or exit Newt's Cape) to close an open connection explicitly.
If there is a non-empty book and NewtPack is installed. The current Newt's Cape book resides in heap and is installed only temporarily as a book. This temporary book disappears if you quit Newt's Cape, reboot your Newton, tap Remove Current Book, or process another document.
If a book with the same ISBN already exists, Newt's Cape prompts you to replace it. NewtPack starts to save the book as a package and displays progress in Newt's Cape status area (Older versions of NewtPack display a "barber pole" and Stop button for NOS 2.x in a separate progress box). The Save Package Process option allows creation of somewhat larger books for 1.x; also see Size Constraints.
During Save, Newt's Cape removes the book to recycle space, but may still run out of heap. If you are using an Original MessagePad (OMP with pre-1.3 ROMs), you should be able to save small packages (~30-40K). If you are planning to use books with Newt's Cape (regular or Lite) present (e.g., these books) or book that make use of explicit http links, each book package will be smaller if you uncheck the Standalone book pkg? preference. A book can also be stored compressed on the Newton to save further space. To create 1.x compatible books with a 2.x Newton, use b&w rather than gray graphics, use portrait rather than full Page Size, use standard (built-in) fonts unless you plan to distribute fonts with your book, and use special 1.x-compatible versions of Newt's Cape and NewtPack (ALIGN on paragraphs with links, and 2.x protos in VALUE_TYPEs should be ignored).
If NewtPack completes successful (you'll hear a sound; otherwise, you'll see an error message; or a close box will appear in old 1.x progress box), you can access the book from the Extras drawer and remove it later as you would any package. If you would like to distribute the book as a package to other users, you can move it to a Mac or Windows desktop using Extract Package (1.x) or PackageBuddy (2.x).
2.x only, if cache on. save current HTML cache entry to Notepad as text. Newt's Cape adds comment with filename as first line of Note. For HTML, it adds BASE before TITLE; very long text source may cause an error
2.x only, if cache on. save current Image cache entry to Notepad as a graphic.
2.1 only, if cache on. save current HTML cache entry to NewtWorks as text.
2.1 only, if cache on. save current Image cache entry to NewtWorks as a graphic.
This feature has been removed from Newt's Cape 2.0, but is accessible via HTMList plugin.
2.1 only. Select a book (from popup) to copy styled text and graphics from entire Newton book (including current book) to a NewtWorks document. In this initial version, any formatting/layout info (e.g., indent, spacing, tables) is lost, form objects are omitted, links don't work. You can also select Save to NewtWorks in book's routing menu if HTMList is installed (which supports page ranges, paragraph justification).
remove the current Newt's Cape book. This should not normally be necessary since the book is removed when you process another document, save a book or exit Newt's Cape. But you can reprocess the same URL, or free up heap sooner by using this.
discard current entry from HTML or Image cache
2.x only. If an error leaves Newt's Cape in a strange state, or a transfer appears hung, or you would like to disconnect NIE without quitting Newt's Cape, you can select this command instead of closing/reopening Newt's Cape -- it may continue with queued requests, i.e., for Load Images. This releases NIE connection and removes any passwords (via authentication). If no other NIE applications are active, depending on your NIE Preferences, you will either immediately disconnect or see a flashing star at top center.
Display the current book. If a book with the same ISBN already exists, Newt's Cape will not display the book -- instead, a Confirmation box will (eventually) appear, prompting you to replace the existing book/package. After removing it, you should be able to view the new book. If book has been removed but URL is visible, this is same as File:Open Location.
retrieves URL, and replaces existing source (if any) in cache, only if successful
If Image Cache is enabled, Newt's Cape scans current cached document for IMG and INPUT TYPE=IMAGE (and META ICON) URLs in current cached document; opens overview of image (GIF,JPEG) URLs, similar to HTML Cache. Tap on a URL to load just a single image; otherwise, select items with checkboxes and select "Load" from action menu to load selected images. After fetching image(s), Newt's Cape reprocesses document to show image(s) embedded. If some images do not load due to server timeout, just do Load with Images again; if some GIF or JPEG images still don't load, send me the document and image URLs so I can test. If the images are large or you get GIF conversion errors (e.g., 50K @ 16-grays), you might try using a lower resolution and check that you have enough space available on your specified Image store.
allows selection of links in current document to load to the cache. (Ideally, this could be done via a special autoload preference and Schedule, but since there are potentially many links in a document, this is a manual experimental option for now). It should includes all http: URLs, but it does not check in advance for helperApps; it could include ftp: URLs if NewtFTP supported getURL API.
If current document is a FRAMESET document, View:Load Frames lists any individual frames that can be loaded.
display (cached) HTML source for an .htm/.html document via Newt's Cape book viewer as PRE text (default). You can also use Save HTML to NewtWorks, or Save HTML to Notes if source is not too large, or specify a different helper app to view "text/plain". After viewing source, you will need to remove the current book in order to browse/parse the HTML for this URL.
2.x only. creates a book with a table listing currently installed MIME types (sorted alphabetically), application symbols and built-in items, and specific file extensions (if any).
return to the previous link location (in history list). same as the Back arrow button in the book control bar.
return to the next link location (in history list) -- after a Back. same as the Forward arrow button in the book control bar.
goes to Other Options:Startup URL, if specified.
processes "next" newer item (URL of item above in HTML Cache overview)
sets starting item in HTML Cache to current URL, if any; otherwise, oldest item in cache; it then does "Newer Item"
processes "next" newer item (URL of item above in Image Cache overview)
sets starting item in Image Cache to current URL, if any; otherwise, oldest item in cache; it then does "Newer Item"
clears history list
a list of up to 20 most recent link locations (most recent at bottom). select an item to go there directly.
2.x only. Newt's Cape obtains the current URL first from the URL area -- if the text begins with http:. You can drag/copy the current URL from the URL area to Notes or to a 3rd party bookmark manager; or add a HyperLink to HyperNewt from the book.
opens overview of current bookmarks, similar to HTML Cache.
adds URL or title to user bookmark soup; if bookmark with same URL already exists, it updates the title and places in Unfiled. bookmark appears in Bookmarks menu and/or overview. You can also add a bookmark from a book link or HTMList.
removes URL from current menu and bookmark soup. also see overview Delete command.
annotated collection of examples: text, form, graphics, package, NewtonScript.
folders for bookmarks in current user bookmark soup; select folder for submenu of individual bookmarks. Unfiled folder normally appears; if Unfiled bookmarks in menu preference is on, Unfiled items appear in initial menu (instead of Unfiled folder).
We now return to covering the many preference settings available in Newt's Cape.
Specify Newt's Cape settings, especially NIE-related. For 1.x, only Doc Done Sound, Page Size, Scale Images to Fit?, Pkg, NewtonScript, and Startup/Proxy URL appear.
2.x only. If you have a number selected (i.e., not "no cache"), Newt's Cape saves any source obtained via NIE (or serial optionally) in a soup called HTMLCache:NewtsCape. You can remove individual entries using the overview File:Open HTML Cache or Remove from Cache.
2.x only. Newt's Cape saves any image (GIF only currently) obtained via NIE in a soup called ImageCache:NewtsCape.
1.x only checkbox: same as 2.x Images option (above); note: PICTs are not scaled.
2.x only. sets the graphic resolution for GIF conversion.
2.x only. Cookies are small amounts of state information provided by certain servers.
If a server suggests that Newt's Cape is not capable of handling cookies, it is probably discriminating on the User-Agent field. One solution is to contact the site administrator and ask them to recognize "User-Agent: Newt's Cape/...". It might also be possible to add a User Agent Prefix, e.g., "Mozilla/2.0 [en]" to try to trick the server into thinking you're using a suitable version of Netscape, for example.
built-in sound to play when document finishes processing (default: none). 2.x includes any additional registered sounds, e.g., www.tactile.com)
select page size to use -- 1.x default: portrait; 2.x default: full
Notes: Actual vertical size is 2 pixels less. A portrait book on landscape 2.1 may not update page number properly until it's saved as a package. You can also specify your own page sizes directly via Book Page Width/Height preferences.
Help books have a smaller screen size (and different viewer), and are best suited for small pages with headings. Links to help sections display just the corresponding heading in the Topics list (not the content page directly). You should set Above for Hx tags to 'Top. Too many headings may not scroll. Since help books differ slightly in underlying implementation from regular books (and 1.x vs. 2.x), there may be other problems (reports and feedback appreciated). Help books can be saved as separate packages or as part of a NewtDevEnv application.
if NewtPack is installed, this picker appears. Note: the settings before processing the document are used (changing just prior to saving does not apply to current book).
Although we are not aware of any actual situations, it is possible that a web page could contain a malicious fragment of NewtonScript. When set to none (default), Newt's Cape ignores any NewtonScript embedded in HTML documents. If you create your own pages, download trusted pages (e.g., examples.htm) or examine the source yourself, you may want to change the setting to Compile to allow Newt's Cape to evaluate expressions and compile methods while processing those documents in order to provide richer appearance and functionality.
if on, automatically hides Newt's Cape when it's finished processing a document; you can reshow Newt's Cape via flashing star at top(2.x), or Newt's Cape button in book control bar
If a proxy or host requires "basic" authentication, it will prompt for user name and password as needed; after completing, tap Authenticate to reload the page. This information is supplied automatically for subsequent requests to the same proxy or host. There may be situations where you want to provide authentication in advance, e.g., you've turned off "trust" on caches and are about to Reload a page or Load with Images and are going through a proxy server that requires authentication or to a host that requires authentication. You may also want to preserve authentication information across multiple sessions. To prevent multiple password prompts from appearing, you can proactively login in advance -- just tap this button, fill in the host (without http://), user name and password. You can also Authenticate in advance for a set of scheduled pages.
This picker contains many miscellaneous options. Items with non-default values are prefixed with a dot. The 3-line input field below is used to display/enter the value. For some items, any non-empty string specifies "true".
After Newt's Cape starts up (or Go:Home), :getURL is called with this URL (this can also be a local URL such as "Notes/foo.htm").
On 1.x, this appears as Proxy Host. In order to access the net through firewalls, you can supply a proxy host and port, e.g., http://www.foo.com:80. For 2.x, it should begin with http://; the port number is optional (default: 80).
You can use a mailto: proxy to send URL requests to a mail server, especially for 1.x, or for 2.x, even if NIE is installed. Since email@example.com is no longer available, try one of the following free web-mail servers below (if you find others, please let me know). Since these may differ in command syntax, the mailto: proxy can include subject (the examples below are empty) and body fields (the body field is a pattern expression with the document URL substituted as a parameter). Examples for some currently known servers:
Note: SpeedyLink requires pre-registration (free) and returns text-only pages; link navigation via special commands.
2.x only. Even if you have specified an http proxy, you can specify that it be used only for certain hosts. Include any domains, separated by semicolons, e.g., foo.com; xyz.org.
2.x only. Even if you have specified an http proxy, you can still access certain hosts directly. Include any domain exceptions, separated by semicolons, e.g., foo.com; xyz.org.
2.x only. You can create a Find options frame for your favorite search engine; the current default is AltaVista(Text). This allows you to highlight a phrase in Notes or NewtWorks, and tap Newt's Cape:Find. There are several fields to supply (best obtained by looking at the HTML form that you use for the search engine): Find.Title, Find.URL, Find.Method, Find.Data (and Find:CheckData to "compile" the frame).
Find.Title appears in Notes menu when a phrase (not URL or HTML source) is selected. e.g., AltaVista would appear as Newt's Cape:Find (AltaVista Text). If this is empty, Notes will not display a :Find command; if Newt's Cape is closed and it's unclear if there is a Find pref, it will display :Find?
URL for a search engine, e.g., http://altavista.digital.com//cgi-bin/query. This would typically be the ACTION of a search form.
To set Hemlock for Find, set Find.Title to Hemlock, so that Newt's Cape:Find(Hemlock) appears (if Newt's Cape is open); no need to change Find.Method or Find.Data. Change Find.URL to Hemlock following by a list of search engine names (as they appear exactly in Hemlock; assuming they've been added/initialized), separated by semicolons; for example,
GET or POST -- corresponds to the METHOD of a search form; must be uppercase. e.g., GET for AltaVista.
NewtonScript source for an array of alternating data slots and values e.g., for AltaVista Text: ["q","","text","yes","pg","q","what","web"]. You would typically obtain these by looking at a FORM and its INPUT, SELECT and TEXTAREA objects in a search page. Note: first slot in the array must be the "query" slot -- Newt's Cape sets this slot with the selected phrase from Notes or NewtWorks. Slots correspond to case-sensitive NAME attributes. Values are strings for default and hidden VALUE attributes; for checkboxes, enclose the value(s) for same name in an array, e.g., ["checked"].
In case of very long data definitions (since this is only a 3-line text field), enter last fields first, and then add other items before these to keep most recent visible; select CheckData several times to catch any errors early (since you won't be able to edit the last items conveniently). If you add this for your favorite search engine, let me know the parameters you used, and I'll summarize for other users.
compiles and checks the data frame definition -- you must do this after editing Data. Possible responses:
2.x only. default: none. primarily for advanced users/testing. some servers may use the User-Agent request header attribute to route you to different pages, or to assume (incorrectly) certain capabilities, e.g., cookies; this allows you to add a prefix to the usual "Newt's Cape..." label. e.g., "Lynx Newt's Cape...".
2.x only. default: none. some experimental web/proxy servers might generate different pages and/or graphics based on browser/screen size and resolution. If non-empty, includes fields in request header:
Accept-Language and Accept are for customizing the HTTP header request so that
certain servers might return a different version of a page.
If you specified
fr; q=1.0, en; q=0.5, for example,
and if server recognized this, server would return a French version of a page; otherwise, English.
Newt's Cape normally sends
if you specified
text/html-newton; q=1.0, text/html; q=0.9, */*,
and if server recognized this, it would return a Newton version of a HTML page;
otherwise, regular HTML or other MIME type.
non-empty value means login automatically (without confirmation) using default link setup
This releases Newt's Cape's NIE connection after interactive transfers, allowing NIE to disconnect after usual timeout. (Connection is normally released automatically after quitting Newt's Cape or finishing Schedule, or a manual File:Reset HTTP)
isbn (default: introtoc.htm) of a local book to follow when selecting i:Help
if on, generally keeps Newt's Cape invisible -- primarily intended when Newt's Cape is accessed from other applications. A safer preference to try is Auto hide? option.
if on, keeps current book open until the next document begins to display. Advantage: keeps content on screen as long as possible; disadvantage: ties up heap.
if on, prevents automatic redirection via META REFRESH tag; hopefully page provides a manual link. Note: this does not prevent server-based redirection
if on, ignores table tags. If certain pages do not display correctly due to hardcoded table layout (e.g., width) information, differing numbers of columns, or table nesting, you may want to try turning on this option and reprocessing document.
If off, Newt's Cape includes only links from main document to individual frames. If on, it temporarily appends content of individual frames into main document (for a more complete merge, see Merge Source)
lets you specify a different tag (and thus appearance) to display "frame heading" (default: H1, so it appears in TOC); when frame is included, frame TITLE or filename is surrounded by parentheses
if non-empty does not use user installed fonts for FONT tag, but always maps to built-in(Simple,Fancy) and PRE fonts.
specifies text to use if missing IMG (with no ALT attribute); if empty (default), "?" icon appears
(if non-empty) displays unknown tags (handy for XML or mutant HTML); otherwise omitted (default)
this is used to customize the behavior of the Newt's Cape (icon) button. if the preference is empty or not recognized, it performs File:Open Location. If the value partially matches Bookmarks, this is the same as if you had selected Bookmarks:Browse/Schedule; if the value partially matches Open Location, Notes, Inbox, Paperback, WabumiUni, NewtWorks, HTML Cache, Image Cache, or Nethopper, the appropriate File command menu appears. It also matches Newer HTML(Image) Cache Item for Go menu. Note: available items depend on your version and language -- see what commands appears when you tap the File button. e.g., Works would match NewtWorks; HTML would match HTML Cache; Eingang would match "Inbox" (in German version). You could still open an occasional URL in status area by tapping File:Open Location.
if on, displays full screen overviews underneath other floating views, such as Button Bar Plus; since it would float underneath main Newt's Cape view also, you would move or hide/reshow Newt's Cape to see bottom-most overview items and file/action buttons
allows you to customize the top row of added keys; the preference value is a list of comma-separated labels which replace corresponding keys: http://,www.,.com,.edu,.net,.htm,.html. You can include a space (or end list prematurely) to skip existing keys; erase value to return to original defaults. Does not allow modification of key widths, so widest keys should be first or second; last key is slightly wider than middle keys. You'll have to manually close old keyboard in order to open new one.
File:Open NewtWorks should provide a complete list of documents that contain <HTML>; if this is too slow and/or the Newton runs out of heap when you select File:Open NewtWorks, you can try turning on this option (non-empty value); if on, Newt's Cape checks only the "summary" slot of the document without actually doing a full text search; however, this apparently can miss some documents -- either that don't have <HTML> in first 80 characters, or that don't have a summary slot at all, e.g., transferred via NCU?; or, you might try making "missing documents" visible by selecting entire content and copying to a new NewtWorks document (or just turn off the preference)
specifies the body text to appear in an email message when you tap a mailto: link. set this to something non-empty like "hi"; if empty, no message is generated/placed in your OutBox; default for new users or after resetting Main Pref Set: "hi, I saw this cool document in Newt's Cape".
a list of substrings (host, file extension, URL fragment) to filter out URLs when scanning for links to automatically load in scheduled documents. e.g., www.foo.com; .doubleclick.com; .sit, .zip. Note: all scheduled sets use this preference in current PrefSet.
There are several options you can set before processing a documents into a book -- be sure to reset manually later (ideally, these would be edited via a book's About dialog, after creation of a single book):
After you register, you can check this to prevent the About box from automatically appearing when you start Newt's Cape.
You can switch between sets of preferences ("Pref Sets") in the main i menu-- this is somewhat like the Macintosh Extension Manager. You can manipulate Pref Sets via the action button. A Pref Set could be useful when using different fonts for a particular language Encoding, or switching to "dumber" settings for problem web sites, e.g., Ignore Tables, don't Include Frames, etc.; all preference-related info, including cookies and authentication, is stored with current Pref Set. Pref Sets appear under "i" menu, with only Main Pref Set initially. When you select a different Pref Set, Newt's Cape does some reinitialization; selecting same set opens General preferences. On the action button in i:General:
Customize properties of different tags, e.g., font, indent, etc. These prefs are saved across sessions as System preferences. (Note: individual documents can override these Prefs via META -- this is discussed in an advanced NewtonScript document).
2.x. If Encoding plugin is installed, along with Japanese or Chinese encodings. select a NewtonTextEncoding for exchanging data between desktop or http server and Newton (default: 'sjis = ShiftJIS). Thanks to Shinobu Yamada; for cool, funny, useful applets or information for Newton users in Japan, see MarbleDevelopment. Thanks also to "GNUE" (aka Makoto Nukui) for the NewtonTextEncoding(NTE) development tool.
You can switch between sets of preferences ("Pref Sets") in the "i" menu -- learn more about creating, renaming, deleting Pref Sets.
Under i. Process displays current processing options (including htmlApps). Note: for NOS 2.x, the current screen orientation (via system Rotate button) can be used to create the book in portrait or landscape (see Page Size).
As Newt's Cape adds a content item to the book, it immediately displays and updates the same book page (unless the book already exists -- see Display). At the end of the document, Newt's Cape positions the book to page 1, # link destination or page from History menu, and optionally plays a sound.
The default. Newt's Cape displays each consecutive finished book page (while the next page is processed). This is much faster than Display Item (especially on NOS 2.x).
Similar to Display Page, except that it does not scroll automatically to subsequent pages during processing.
Newt's Cape displays the book only when it is entirely finished. For documents that contain many animated graphics, especially on slow or heap-challenged MessagePads, this option may be best.
For NOS 2.x and NIE, this can fetch an HTML, GIF or JPEG document into cache, but without parsing and displaying it. You can process it later, e.g., when logged out to save time or maximize heap. This is also handy if you want to experiment with the Gray Levels or B&W threshold for unconvertedGIFs.
if NewtPack is installed. It is similar to Display All, followed by selecting Save as Package (under File). However, with this option, Newt's Cape can create a much larger book (especially on 1.x) -- since it does not have to actually display the book, it can store some objects (text and graphics) in a separate temporary area on your default store rather than in limited heap.
2.1: creates a Newtworks document directly from HTML, rather than create a book first and then use File:Save Book to NewtWorks. This is much faster; however, it has the same problems as Save Book to NewtWorks: long lines not wrapped (horizontal scrolling needed), horizontal layout (e.g., tables) not supported, links inactive, and form objects not included. URL appears at end of document.
Other applications can install process options. For example, an application might collect bulleted items into bookmarks or into a Notes outline. see installHTMLApp API and htmlApp example.
if bottom triangle expanded. The status line (below commands) displays information for NIE*/serial connection and transfer progress, parsing (number of tags/current tag), installing a package, and saving a book (NewtPack). It replaces various progress boxes in earlier versions. Status information is refreshed as time permits. since this is based on lines currently, really l-o-n-g (many K) lines, e.g., query results with few line breaks, may appear to hang (after completing)-- just wait awhile for line to finish parsing. Status gauge may appear on right side (above Stop) to indicate transfer, parsing or installation progress.
*If you do not have an active NIE connection, the NIE connection box appears the first time you attempt to access a URL that's not in the cache. This document assumes that you already know how to configure and establish your connection. If you close the connection box (rather than tap Connect), or tap Stop during the connection process, Newt's Cape should stop accessing the current URL (and also flush any other pending requests, e.g., graphics). If a site requires a name and password, an authentatication dialog appears.
if bottom triangle expanded. The error line (below status) displays error messages from transfer or parsing. If you are using a serial or ADSP connection, error messages may be printed to your terminal emulator.
Sometimes you may want to get Newt's Cape out of the way without dragging, or a more time-consuming close/reopen. To hide Newt's Cape, tap the small star(2.x)/gray close box(1.x) to the left of the close box. You can also set Auto hide preference so that Newt's Cape automatically hides after processing current book. You can reshow it quickly from the flashing star at top center of screen (2.x), book bar(Newt's Cape icon) or Extras.
When Newt's Cape quits, it releases any open connection (NIE or serial), removes current temporary book, restores heap, and cleans up the 2.x caches (checks/removes any temporary or old entries to maintain number of items specified for HTML and Images caches. it also discards any authentication information). If you just want to hide the Newt's Cape window temporarily, use the star to its left.
If you are using a "freeze manager", you can minimize problems by manually Closing any open serial connections and doing Reset HTTP (any NIE connection) and waiting a few seconds before quitting. (Ideally, freezing managers would allow a few seconds for asynchronous closing of communications)
Apparently, the 2.1 book reader maintains a large number of entries for "offline book" entries (for annotation and bookmark info); Newt's Cape now removes all except the most 20 recent offline entries (rather than let this grow indefinitely).
The book consists of several areas:
During processing (if Display Item or Display Page is selected), or after selecting View:Current Book in Newt's Cape or tapping the book's icon in the Extras Drawer, you should find yourself in the current Newton book.
While the current document is still processing, you should not follow links until the document is finished. If there are a lot of animated graphics on a page, you may want to try to scroll to a different page to minimize contention (or use Process:Display All). If you need to abort the current document (e.g., too large or strange), you can try the Stop button. A document might be truncated if a server disconnects prematurely, or sends extremely long lines which overflow buffers.
You have some control over the appearance of your document: fonts, page breaks, etc. (i:Appearance).
If a ? icon appears for missing graphics, you can load graphics for current document by using a line gesture, or selecting View:Load with Images. If a graphic appears with a 2-D scroller at its lower-left corner, you can re-parse the document after changing scale to fit. Underlined text, outlined graphics usually indicates a hypertext link or image map. In addition to tapping a link, you can use several gestures.
If you're not sure what's behind a link, it's best to use horizontal gesture rather than just tap. A link may appear not to work if there are bold or unusual font changes earlier in the same paragraph. If a URL matches a local object, that application or book opens (for NOS 2.x, if local document is a frozen book or application, it is thawed). If the destination is an absolute URL (begins with http://...) or is relative to the BASE of the current document, Newt's Cape uses NIE (for 2.x) to request the document; for 1.x, it generates a WebMail request (you can delete these from your Outbox if you change your mind).
This bottom bar should already be familiar to you (other books, MessagePad manual); Newt's Cape adds Newt's Cape icon, back and forward buttons.
These are usually found on the silkscreen or floating menu bar.
You can do three things from NewtWorks (2.1 only) or Notepad documents:
In NewtWorks, select text in a Paper (not Drawing) document; or, in Notes on any MessagePad. If no selection, entire document is used (except 2.x Outline/Checklist). Note: if Newt's Cape does not appear in Tools menu, you should make sure that NewtWorks installs before Newt's Cape, perhaps by storing NewtWorks on Internal; or, you can freeze/thaw Newt's Cape to force it to register with NewtWorks manually. Here is how each action is recognized/performed:
Save as HTML command also appears if HTMList is installed.
Similar to Notepad and NewtWorks, you can process a list of URLs (:getURL), or process an HTML source document (:Open) in the Inbox.
Similar to Notepad and NewtWorks, you can process a URL (:getURL) that is in a custom "URL" field, or in the Notes field of Names (it ignores any <!-- HTML comments --> in Notes), or process an HTML source document (:Open) in Notes.
This document (in all its formats) is © 1995-2007. Steve Weyer, Greg Simon. All Rights Reserved Worldwide
Version 2.1. Last updated: Dec 2000