labels.facet_contentLength_5m=5MB -  
labels.facet_filetype_title=文件类型
labels.facet_filetype_html=HTML
labels.facet_filetype_word=MS Word
labels.facet_filetype_excel=MS Excel
labels.facet_filetype_powerpoint=MS PowerPoint
labels.facet_filetype_odt=ODF Word
labels.facet_filetype_ods=ODF Spreadsheet
labels.facet_filetype_odp=ODF Presentation
labels.facet_filetype_pdf=PDF
labels.facet_filetype_fb2=FictionBook

Emoji","Segoe UI Symbol","Noto Color Emoji";font-style:normal;font-weight:400;line-height:1.5;text-align:left;text-align:start;text-decoration:none;text-shadow:none;text-transform:none;letter-spacing:normal;word-break:normal;white-space:normal;word-spacing:normal;line-break:auto;font-size:.875rem;word-wrap:break-word;opacity:0}.tooltip.show{opacity:.9}.tooltip .arrow{position:absolute;display:block;width:.8rem;height:.4rem}.tooltip .arrow::before{position:absolute;content:"";border-color:transparent;border-s...

Because this code is executed before the application **starts** taking requests, and right after it **finishes** handling requests, it covers the whole application **lifespan** (the word "lifespan" will be important in a second 😉).

## Server Machine and Server Program { #server-machine-and-server-program }

There's a small detail about names to keep in mind. 💡

The word "**server**" is commonly used to refer to both the remote/cloud computer (the physical or virtual machine) and also the program that is running on that machine (e.g. Uvicorn).

        writeInt2(andxOffset, dst, start + ANDX_OFFSET_OFFSET);

        andx.useUnicode = useUnicode;
        if (andx instanceof AndXServerMessageBlock) {

            /*
             * A word about communicating header info to andx smbs
             *
             * This is where we recursively invoke the provided andx smb
             * object to write it's parameter words and bytes to our outgoing

     */
    protected int writeString(final String str, final byte[] dst, int dstIndex, final boolean unicode) {
        final int start = dstIndex;
        if (unicode) {
            // Unicode requires word alignment
            if ((dstIndex - this.headerStart) % 2 != 0) {
                dst[dstIndex++] = (byte) '\0';
            }
            System.arraycopy(Strings.getUNIBytes(str), 0, dst, dstIndex, str.length() * 2);

			recordDelimiter: "\n",
			fieldDelimiter:  ",",
			sendErr:         nil,
			header:          true,
			wantColumns:     []string{"header1", "header2", "header3"},
			wantFields:      "ok1,ok2,ok3\n" + `"a ""word""",b` + "\n",
			wantErr:         io.EOF,
		},
		{
			// This works since LazyQuotes is true:
			file:            "invalid-baddoubleq.csv",
			recordDelimiter: "\n",
			fieldDelimiter:  ",",

    Comparator<byte[]> comparator2 = UnsignedBytes.lexicographicalComparatorJavaImpl();

    /*
     * These arrays are set up to test that the comparator compares bytes within a word in the
     * correct order—in order words, that it doesn't mix up big-endian and little-endian. The first
     * array has a smaller element at one index, and then the second array has a smaller element at
     * the next.

        // UID (2 bytes)
        buffer[28] = 0x03;
        buffer[29] = 0x00;

        // MID (2 bytes)
        buffer[30] = 0x04;
        buffer[31] = 0x00;

        // Word count
        buffer[32] = 10; // 10 words = 20 bytes

        // Parameter words for transaction response
        // Total parameter count (2 bytes)
        buffer[33] = 0x00;
        buffer[34] = 0x00;

The generated code still has some **duplicated information**.

We already know that this method is related to the **items** because that word is in the `ItemsService` (taken from the tag), but we still have the tag name prefixed in the method name too. 😕

We will probably still want to keep it for OpenAPI in general, as that will ensure that the operation IDs are **unique**.