If you find blocks confusing, it may be because people often use the term block to refer to two different concepts: a block definition and a block reference. A block definition is the equivalent of a stamp with your name on it that you keep in your desk drawer. It exists, but your name isn't yet stamped on anything. Once you take out the stamp, ink it, and press it onto a sheet of paper, you have the equivalent of a block reference.

WBLOCK isn't a block command; it's a save command. I've fielded countless questions over the years stemming from confusion over this command. I wish it had been named something with the word save in the name, but it wasn't. Just keep in mind that you're creating a new drawing file whenever you use this command, and it isn't necessary to use this command to create a file that you intend to use as a symbol in other drawings. You can use any drawing file as a symbol.

The great utility of the WBLOCK command is the flexibility with which you can create drawing files from part of an existing drawing. Let's say you want to create a separate drawing file using only a portion of an existing drawing file. There's no need to erase objects you don't want in the new drawing—just use the WBLOCK command. There are three ways to use the WBLOCK command to partial-save a drawing (hey, PSAVE might have been a good name). Let's look at each one:

Use any portion of a drawing to create a new drawing file by selecting entities directly and identifying an insertion point. Using the command-line version of WBLOCK, press the Enter key when prompted for a block name. Now entities can be selected in the same way they are for the BLOCK command. When you use the command-line version of WBLOCK, the selected entities are erased. To get them back, use the OOPS command.

INSERT

The INSERT command has a useful dialog box, of course. It allows you to view a list of all block definitions, or other DWG files, and select one you want to insert. But you can also use INSERT at the command line by placing a minus sign in front of it when typing: -INSERT. This is helpful in creating AutoLISP programs that insert a drawing into another drawing.

Figure 5.2. Using -INSERT to see the current path

There's another, more unusual use for the command-line version. Typing -INSERT followed by a nonexistent block name gives you a quick indication of the search path AutoCAD is currently using, because it lists all the places it looked for the nonexistent block definition you specified, in order. The first place AutoCAD always looks for a block definition is in the drawing itself. The rest of the list contains the directory where the drawing is currently saved and then the AutoCAD search path specified in OPTIONS. See Figure 5.2.

Note

If you try to insert a drawing using the command-line version of INSERT, you may not get what you expect. If the filename you specify is the same as an existing block definition, the existing definition is used instead of the drawing you specify. Unlike the warning you get with the INSERT dialog box, AutoCAD uses the existing definition without asking if you really mean it.

A brand-new space first appeared in AutoCAD 2006. In addition to Model Space and Paper Space, we now have what I think of as Block Space. Block Space is an editor that makes defining and editing blocks much easier, because you can use all the standard AutoCAD editing and object creation commands, and the odd-colored background never lets you forget which space you're in.

If you can't stand using the block editor, you can still use the standard Block Definition dialog box to create or edit blocks, or the REFEDIT command. Double-clicking a dynamic block won't bring up REFEDIT by default, but you can change that, if you want.

One thing I've always appreciated about Autodesk is that legacy behavior doesn't usually change in new releases of AutoCAD. This reduces, but doesn't eliminate, the surprises that can confront you went upgrading to a new release. Even though there are now two other tools for editing blocks, the old fashioned method still works just fine.

For years you've updated blocks using the BLOCK command. You still can. Just draw a new symbol, use the BLOCK command, and define a new block with the same name as an existing block. It may save time to insert the existing definition, explode it, and then edit the resulting entities. The redefined block immediately updates any block references that have already been inserted into the drawing. This allows you to globally edit even a scaled or MINSERTed block. Redefining a block in this way affects only one drawing file.

You can also redefine a block by replacing its definition with that of an existing drawing file using the -INSERT command. When you're prompted for the block name, type oldblock=newblock. This replaces an existing definition named oldblock with a drawing named newblock.dwg. Don't forget to include the path along with the drawing name.

If your old block definition and the new block definition have the same name, you can redefine the existing block using the Insert dialog box. If you try to insert a drawing with the same name as a block that exists in the drawing, you're asked whether you want to redefine the old block. Say yes, and all block references are redefined.

When you create a block definition using entities that reside on layer 0, those entities take on the properties of the layer onto which they're inserted. They also return to layer 0 if the block is exploded. The entities that reside on any other layer always have the properties of that layer, even when inserted onto a different layer. This creates a block whose visibility is controlled by more than one layer: the parent layers and the host layer.

AutoCAD ships with many dynamic blocks already referenced onto Tool Palettes. You should start by inserting one of them and then double-clicking it to enter the new block editor. All the things you see floating around the block geometry are objects representing parameters or actions. Because they're objects, you can select them and view or change their properties in the same Properties palette as other objects. Figure 5.3 shows a stretch action.

Because it's so new and so different, let's go over the basics of creating these blocks with just enough information to get you experimenting.

Types of Objects: Actions and Parameters

Actions are things that happen and are usually represented by command names; parameters are the values that control those actions. In Figure 5.3, the window block can be stretched to change the window's height. The action Stretch is controlled (or constrained, to be more inventor-like) by the Linear parameter window height. To create this image, I added two objects to the window in the block editor: a Linear parameter and an action that's attached to that parameter.

Figure 5.3. Dynamic blocks in the block editor

Block symbols should be created actual size most of the time. In other words, a block that represents a washing machine should be drawn the size of a specific washing machine. A schematic block—like a resistor symbol or a duplex receptacle symbol—should also be drawn full-size, but that means something a little different. Because the same symbol is used to represent objects whose actual size differs, draw the schematic the size you want it to measure when it is plotted. If, for example, you want a duplex receptacle symbol to be .25″ when plotted, define the block using objects that are .25″. When you reference the block on a floor plan, scale it to the reciprocal of the plot scale used. It will be the correct size when plotted.

Sometimes it makes sense to use what is known as unit sizing, but a lot of users overdo it. Unit sizing involves defining each block to measure 1 unit × 1 unit. When inserted, the block is scaled in the X, Y, and Z axes to its actual size. In theory, this allows you to use a single block definition for, say, many different sizes of refrigerators. Some symbols lend themselves to unit sizing—a simple door symbol, for example—but most don't, because elements within the symbol get distorted when the scales are different for the different axes.

Here's one of the few uses I have for AutoCAD's SNAP and GRID commands. When I use a snap, I always set a grid that's twice as large. It's a lot easier on the eyes. For schematic symbols, I set a logical snap increment like 0.1 to draw each symbol, making sure insertion points land on a snap. Once I've drawn all the symbols for a schematic diagram using the logical snap increment, I set the same snap increment when I draw a schematic diagram that uses the symbols. I plug them into a grid and connect them with lines that end exactly where I want them—a little like using Lego building blocks. See Figure 5.4.

Once you've defined standard blocks, place references to them on a tool palette as described in Chapter 3. You may want to create a reference drawing by inserting block references into a grid and adding a dot (DONUT) at each insertion point. Include the name of the block, and print the drawing. Now it can be used as a reference in a CAD standards binder. Although it's useful to mark an insertion point with a dot, don't make the dot a part of the block definition. Figure 5.5 shows a partial example of an American Society of Mechanical Engineers (ASME) symbol library.

Figure 5.4. Setting a snap/grid to create schematic symbols

Figure 5.5. Plotted reference sheet for symbols

Related Blocks Grouped into Library Drawings

My recommended method for maintaining symbols libraries involves creating blocks grouped into library drawings. This is the practice that I find is most common. Place all related block definitions in separate drawings that contain only block definitions. If you use a large number of symbols, break them into categories and create a DWG file for each category. You can then use DesignCenter to create a tool palette that contains all the block definitions from each symbol-library drawing or to load individual blocks from the symbol drawing into your current drawing as shown in Figure 5.6.

If you want to bring all block definitions from one drawing into another, insert the entire drawing and press the Esc key when prompted for an insertion point. All the block definitions (and all other user-defined elements) become part of the drawing. This can result in a lot of unused block definitions, layers, dimension styles, text styles, and so on. If so, you can use the PURGE command to eliminate the unused objects.

There's another serious drawback to this method. If you have a naming conflict between any element of the host drawing and the drawing being inserted, the host drawing wins.

This causes a problem if you use the same name for different block definitions in the two drawings. Your changes are discarded if the host drawing contains different definitions. For that reason, I don't recommend inserting a drawing to get the block definitions. DesignCenter is a more efficient method, and you can see what the blocks look like before you insert them.

Note

If you have block definitions in a drawing that was created with an early release of AutoCAD, the blocks may not all display in DesignCenter. Icons representing blocks weren't added to drawing files before DesignCenter appeared. Even if you've edited the drawing in a release that contains DesignCenter, block icons appear only for blocks defined in the newer release (see Figure 5.7). To create icons, open the old drawing, and run the BLOCKICON command. Select all blocks by accepting the default when prompted to Enter block names <*>.

Figure 5.6. Displaying block definitions in DesignCenter

Figure 5.7. Missing block icons in DesignCenter

Attributes can be defined to either appear as part of the drawing or be invisible. Either way, the information attached to a block can be extracted from the drawing into a file for use in another application, generally a database program like Access. The file into which you extract the attribute values can be an ASCII text file or an MDB file (Microsoft Data-base file). Attributes can also be extracted directly into an AutoCAD table.

The commands listed in Table 5.1 are used to create and modify attributes (consider making aliases for them in your ACAD.pgp file). Those commands that begin with the letter E (for extended) are used to manage and extract block attributes manually. Those that begin with the letter A can be used manually but can also be used in scripts and Lisp routines when you're automating procedures that involve attributes.

Attribute definitions are AutoCAD entities; they can be selected and used to create blocks, either by themselves or in combination with other entities. Usually they're used to annotate graphic symbols or complete title blocks, so they're generally used with other entities. Follow these steps to use them in blocks that also contain geometric entities (See Figure 5.10):

When you're defining attributes, you have the option of setting their mode to Invisible, Constant, Verify, or Preset. The modes affect attributes in the following way:

You have several options when defining attributes. The options include:

When you insert the block, the Edit Attributes dialog appears, allowing you to fill in the values you want. If it doesn't appear, you may have ATTDIA turned off, in which case you'll be prompted at the command line. If you get no dialog box and no command-line prompt, you may have ATTREQ turned off. The Edit Attributes dialog in Figure 5.13 shows the tag Species with the value Quercus rubra.

Whether the values you place are visible on the screen depends on whether you defined attributes that are invisible. Most attributes for things like species, size, and cost are invisible—you want the information, but you don't want it shown on the drawing. One exception is a title block. When you create your title block, the attributes should be visible; otherwise, you won't be able to read them in the drawing. If you want to see and/or plot invisible attributes, you can set the variable ATTDISP to ON, as in Figure 5.14.

Figure 5.13. Edit Attributes dialog

Figure 5.14. Displaying invisible attributes

Attribute values can be useful even if they are only used to display information in a drawing, but they are most powerful when you use them to attach information about each block reference. Most of that information will be invisible, so you have to be able to extract it from the drawing in order to use it. That process can be done in one of two ways. I'll go through both. One is easier when used manually, the other useful for extracting attribute values using an AutoLISP program.

There are two methods for extracting attribute information: EATTEXT and ATTEXT.

The EATTEXT command displays the Attribute Extraction Wizard. It's easy to use, and with it you can extract attribute data from a group of selected block entities, either from the current drawing or from any saved drawing or set of drawings that have been referenced into a sheet set. (See Figure 5.15.)

Figure 5.15. Attribute Extraction Wizard

Once you identify where you're going to get the data, you can extract it into one of several different formats, as shown in Table 5.2

The other method for extracting attributes, the ATTEXT command or its command-line equivalent, predates the EATTEXT wizard and is much more involved. I still consider it useful, though, because it can be used with a script or an AutoLISP program to automate the extraction of attributes from multiple drawings. (See Chapter 8.) ATTEXT involves a number of steps.

Before extracting attribute information from a drawing, you have to create a template file to tell the software how the information should be extracted. This template file must be an ASCII file. It can be created in any word processor, but because you must save the file in ASCII format, I recommend using Notepad. EDIT is a DOS alternative that's run from the system command prompt, but its only advantage over Notepad used to be its line counter. (EDIT was the replacement for the dreadful EDLIN text editor. If you remember using that, you can definitely claim the title of "computer old-timer.") If you're using the XP version of NOTEPAD, the View pull-down menu gives you the option of viewing a line counter. There are also other text editors that you can find as shareware or buy. You can even use VLISP, if you find it helpful.

Don't leave any blank spaces or blank lines in this template file. When you try to extract attribute values, if you get the error message "invalid field specification" when you extract the data, you may have a blank line at the end of the file. Remove any extra lines by moving your cursor to the end of the file and using the Backspace key to delete them. Don't use the cursor keys, because the blank lines will still be there. The cursor should end up at the beginning of the blank line that immediately follows your text. Before saving the template file, use the Down Arrow to move your cursor to the end of the file. If the cursor doesn't stop directly below the last line, remove the blank lines, and check each line to make sure you have no spaces at the end of any individual line.

In the template file shown, the first two lines represent attributes that all block references have; they aren't tags that you can create. All blocks have a block name, layer name, a linetype, the coordinates of its insertion point, and other information. If a block includes at least one custom attribute, you can also extract the general attributes. If you use the EATTEXT wizard, you can extract general data from any block, regardless of whether it has any user-defined tags.

Figure 5.16. Extraction template file

The BLANK1 tag creates a space between a right-justified numerical field and a left-justified character field, not because it's named BLANK, but because no values are assigned to that tag. Any false tag name can be used to create spaces, but only once. This file must have a .txt extension and be an ASCII text file. The file appears as shown in Figure 5.16 if you highlight the entire file to make sure it has no blank lines.

An explanation of the file is shown in Table 5.3.

This template file list provides six rows of information in an extracted file: the block name, layer, species, size, cost, and supplier. The second column in the template file specifies the kind of data (characters or numbers), the field width (008 = 8 characters), and numerical precision (002 = 2 decimal places). All character-based fields have a precision of 000.

Once you've created a template file, open the drawing that contains attributes and use ATTEXT or EATTEXT to extract them. You can specify a filename for the extracted data, or let AutoCAD use the drawing name. In this case the result of extracting block information from two tree symbols using this template file is another text file like the following.

What appears in bold is part of the text file created by AutoCAD. The first example is comma delimited:

'att-deciduous','trees','Quercus rubra','2', 120.45,'','Urquhart Gardens'
'att-conifer','trees','Larix americana','3', 200.00,'',' Urquhart Gardens '

The second example is space delimited:

att-deciduous  trees   Quercus rubra     2       120.45    Urquhart Gardens
att-conifer   trees   Larix americana  3        200.00    Urquhart Gardens

Once the information is extracted into a text file, you can insert it into the drawing as text or import it into a database program like Microsoft Access so you can organize the information in any way that's useful. When you use the ATTEXT command, you must decide how the information identified in the template will be listed in the extraction file. The choices are comma delimited (CDF), space delimited (SDF), and Drawing Exchange Format (DXX). DXX is a partial DXF file, containing only block and attribute information. The .dxx extension is used to differentiate the file from a standard DXF file. It has no value to most users. Use SDF for information you want to be able to read or import into AutoCAD. Use CDF if you plan to use the data in a database application. See Figure 5.17.

When an XRef is attached to a drawing, AutoCAD differentiates the layers, blocks, and dimension styles in the XRef from those in the host drawing. To distinguish XRef elements from the host-drawing elements, their names include the XRef drawing name as a prefix. A layer with the name Cen in an externally referenced drawing named Smith_residence.dwg is displayed in the host drawing as Smith_residence|Cen.

Use the Bind option. Otherwise, you may find that your text styles, dimension styles, and block definitions that use the same name as their counterparts in the host drawing have been redefined. Once you use the Bind-type option of Bind, you'll have some fairly long and cumbersome names for layers, block definitions, text style, and so on. You can quickly rename them by using the RENAME command and applying a wildcard (*) to select all the XRef names. For example, if the names of these elements are prefixed with Smith_residence$0$, you can type Smith_residence$0$* for the old name and a new prefix followed by the wildcard—SR*, for example. Figure 5.21 shows the original bind names, and Figure 5.22 shows the result of using the wildcard to rename them.

When you first attach an XRef, the layer visibility of the XRef is determined by its visibility at the time the XRef was last saved. However, you can alter the visibility of XRef layers from within the host drawing to control the display of the referenced drawing. VISRETAIN controls layer visibility the next time you open the host drawing. Will the host drawing display the layers as they appeared when the XRef was saved or when the host drawing was saved?

Figure 5.21. Bind-type bind layer names

Figure 5.22. Use of a wildcard to rename layers

If VISRETAIN is set to 0, changes to layer visibility in the host drawing are temporary, and XRef layer visibility reverts to the XRef drawing when the host drawing is reopened. If VISRETAIN is set to 1 (the default), layer settings are saved with the host drawing and persist from session to session. Normally, you should leave this setting at 1.

When a large XRef is brought into a host drawing, it often contains far more geometry than the drawing requires. To prevent unwanted geometry from being displayed, you can use the XCLIP command to clip an external reference into either a rectangle or a polygon. If you want the frame of the clipped area to show, set the XCLIPFRAME system variable to 1; if you don't want it to show, set this variable to 0.

When a drawing contains many external references or a very large external reference, performance during editing slows down. To reduce the amount of computer resources necessary to manage the host drawing, two indexes can be saved with a host drawing and used during demand loading of external references. These indexes keep track of which XRef-dependent layers are currently thawed and how much of an XRef is being displayed as a result of the XCLIP command. These indexes are added to the database of the host drawing, making it slightly larger, but the performance gain in not loading all layers and all geometry (spatial indexing controls the appearance of geometry) generally makes up for that slightly larger file size.

The INDEXCTL variable controls whether indexes are saved with a host drawing. Its settings are shown in Table 5.6.

You don't need to remember this variable when working with host drawings. All this can be controlled in the SAVEAS dialog box by selecting Tools and then Options. The resulting dialog box allows you to select an index option.

Once you've created and saved an index with your host drawing, you have to enable demand-loading before you gain any performance from setting INDEXCTL. To enable demand loading, either use the variable XLOADCTL or use the Open And Save tab of the Options dialog box, shown in Figure 5.23. Demand Load Xrefs → Enabled With Copy is generally the fastest method when you're working on a network. Only the layers and area specified by INDEXCTL are loaded when a drawing is opened that contains XRefs and the referenced drawings aren't locked.

Although it can be used to edit blocks, the intended application of the REFEDIT command is to edit external references from within a host drawing. This is a potentially very dangerous command, because any changes you make while using it can be saved to the original external reference. Because double-clicking an XRef opens the block editor, it's possible to edit an XRef without noticing it. If you want to prevent the use of REFEDIT on a particular external reference, open that reference drawing and set XEDIT to 0 before saving it. Now you'll have to open the drawing before making changes—a safer method for editing reference drawings, in my opinion.

Figure 5.23. Setting demandloading and index control